These rules and procedures apply to all Trinity College Fire-Fighting Home Robot (TCFFHRC) competitions.
Answers to Frequently Asked Questions will be found on the Contest Website at http://www.trincoll.edu/events/robot/.
The PDF version of this document at http://www.trincoll.edu/events/robot/ should be regarded as definitive; text and font conversion errors may affect other file formats. The HTML version does not include clickable cross-references. Note that the chapter and section numbers have changed from earlier versions and will likely change in the future.
If you find an error or inconsistency, please email the Editor (Ed Nisley ed.nisley@ieee.org) with a copy to the Contest Director (Dave Ahlgren david.ahlgren@trincoll.edu). We will defer problems reported after noon on the Monday preceding the Contest weekend until the next year’s Contest.
Anyone may enter a robot.
There is no limit on team size.
In the rest of this document, the term “team” means either the group or the individual associated with a robot entered in the contest.
The challenge presented by the Trinity College Fire-Fighting Home Robot Contest (TCFFHRC) and the associated regional contests is for contestants to prepare a unique robot of their own design. However, we recognize that some teams may wish to enter a kit-based robot, a commercial robot, or a robot that shares many design features with another robot entered in the contest. Therefore, we award prizes in the kit and unique robot categories in the Junior, High-School, and Senior Divisions. The Walking, Expert, and Assistive Divisions do not have separate kit/unique categories.
A team may enter more than one robot. In order to qualify for a unique-robot prize, however, each robot must differ visibly and significantly from other robots in at least some aspects of electronics or mechanics. Thus an individual, team or school may not register multiple identical robots as separate entries in the same Division except in the kit category.
Multiple, possibly identical, robots that function as a swarm may be entered in the Trinity College Expert Robot Contest. Those robots may not be entered as separate robots in other Divisions.
Each team must indicate whether their robot is a kit robot or a unique robot, with characteristics as listed below, when registering it for the contest.
In cases 2 and 3 above, both of the similar robots will be considered as kit robots.
Registration for the TCFFHRC is available only on-line through the contest website. We will accept registration applications from 12:01 a.m. on January 15, 2012 to 11:59 p.m. on March 20, 2012. For further details check the Contest Website at http://www.trincoll.edu/events/robot/.
The steps in the registration process are as follows:
You must register for the contest between January 15 and March 20 (midnight to midnight), otherwise your robot will not be in the contest. There are no exceptions.
You have spent hundreds of hours and perhaps as much money on your robot. Register early!
A non-refundable registration fee is required for each robot entered into the contest. The fee must accompany each entry.
If you want to enter two robots, then you must build two robots: the same physical robot cannot be entered twice, even if two entry fees are paid.
We repeat: registration fees are non-refundable.
The Division fees for 2012 are:
The Division structure makes the event more enjoyable for students, but it opens up an area of possible conflict. The problem occurs when a group consisting of people both in and out of school enters a robot in the Junior or High School Division.
The Judges will decide whether a given robot has been entered in the correct Division based on both the robot’s capabilities and the team’s abilities.
For example, consider a second-grade student who enters a microprocessor controlled, stepper motor driven robot that uses modulated IR sensing and a video navigation system. The control program seems to be written in C++ and the student’s parents just happen to work for NASA.
This robot would be reclassified in the Senior Division.
Normally a robot created by a group of 6th and 7th grade students with an adult advisor would enter the Junior Division. Such students may also have the ability and skill to build and program the robots that may be entered into the Junior and High School Divisions.
This does not mean that the students have to do everything, i.e., mechanics, hardware, electronics, software completely on their own. On the other hand, we do not want to see an advisor spending hours upon hours writing and debugging a student’s software. We are less concerned about the role of an adult who helps a team of college students since the team would enter the Senior Division, which is open to everyone.
Adults helping students is perfectly OK; that’s how students learn.
Adults taking over the project is not; that’s how student learn to cheat.
As far as the students are concerned, the goal of the contest should be education, not winning. We know that the students (sometimes desperately) want to win, but their adult mentors must allow them to compete, win or lose, on their own merits.
This contest runs on the honor system, but we expect that the student contestants bear primary responsibility for their robots. Should we find any case to the contrary, we will reassign the robot to a more appropriate Division. In these cases the decision of the Chief Judge is final.
Teams should build their robots and bring them to the contest ready to compete: this is not a construction contest where you build robots at the event!
Trinity will provide limited time and space for last minute changes, adjustments, and improvements. However, the robots should be completed (or very nearly so) by the time they get here.
We have eliminated qualification trials. Every team registered for the contest will have the chance to run their robot on Sunday.
However, we have adopted a rule in the High School and Senior Divisions to eliminate unreliable robots:
Thus, the first two rounds of the competition serve as elimination rounds for those Divisions.
TCFFHRC events will be held at Trinity College in Hartford, Connecticut, USA, from Friday 30 March 2012 through Sunday 1 April 2012.
The full schedule of events for the contest weekend will be posted on the Contest Website at http://www.trincoll.edu/events/robot/.
The rules and information in this Chapter apply to all Trinity College robot contests, unless otherwise noted.
The Chief Judge is the final and absolute authority on the interpretation of all rules and decisions.
A team may challenge any ruling or scoring of the Arena Judges by stating that they wish to appeal the problem to the Chief Judge. The Chief Judge will then be called in to decide the matter.
The challenge must be made before the team leaves the arena after the completion of a trial.
All results, scores, and decisions become irrevocable after the team leaves the arena.
Any Contest official may stop any robot at any time if, in their opinion, it is performing or is about to perform any action that could be dangerous or hazardous to people, facilities, or other equipment.
Robots must not use flammable or explosive materials to extinguish the flame.
The goal of the contest is to make a robot that can operate successfully in the real world, not just in the laboratory. Such a robot must be able to operate successfully where there is uncertainty and imprecision, not just under ideal conditions. Therefore, the arena dimensions and other specifications listed below will not be precisely what the robots will encounter at the contest: they are provided as general aids.
The size limits on robots are, however, absolute and will be enforced by the Judges.
Object dimensions are generally given as length x width x height, as the robot encounters the object.
In the RoboWaiter Contest, “deep” refers to the front-to-back dimension of shelves.
Although the robot contest arenas present an idealized version of the real world, you must not assume that all corners are exactly square, all walls precisely vertical, all joints flush, all fasteners recessed, all surfaces uniformly colored, and so forth and so on.
Every robot must successfully handle small misalignments, inaccuracies, discolorations, and other arena imperfections. You must test your robot under less-than-ideal conditions and verify that it works properly.
The contest takes place in a gymnasium that will be quite different from your classroom, laboratory, basement, or living room. Some possible problems you should consider:
The Contest officials will provide a Robot Checkout Table where you can verify that your robot meets various specifications:
An official will assist you and explain any problems.
Each robot has an assigned number that determines the order in which it will compete in the contest. Robots make trial runs in the arena in ascending numeric order, so that the robots compete consecutively. When all robots have completed the first trial, the sequence repeats for the second and third attempts. Once assigned, the order of running will not be changed.
Contestants will have limited time between their trials for adjustments, modifications, and repairs to their robot. However, after the preceding robot has completed its trial, then your robot must be in the arena and ready to start within 1 minute. The Judges will start a timer when they call for the next robot: that robot must be on the Judge’s Table before that clock reaches 1 minute. Any robot that is not ready to compete after 1 minute will forfeit its chance at that trial. It may still compete in any remaining trials.
Translation: If you are not ready, you miss your turn. The End!
The team will place the robot on the Judge’s table and give the Judge the Trial Run Checklist when they arrive for their robot’s trial.
The team must not transfer any information to the robot regarding the layout of the arena, the starting position, or the position of any objects.
Team members must not touch the robot after presenting it to the Judge.
The team will show a Judge how to start the robot by:
After the robot is ready and the Judge knows how to start it, the Judges will determine the location of any objects within the arena, as determined by the robot’s Operating Modes. The Judges will then place the robot and the objects in the arena.
The Judge will determine when the trial begins and will start the robot.
If for any reason the robot does not start, then that trial is over.
All robots, including those using Sound Activated Mode, must have exactly one Start Button switch that starts the robot.
This button must have these characteristics:
Figure 2.1 shows sample Start Buttons. Note that you must provide a green background even if the switch is located on a green circuit board.
You should verify that your robot’s Start Button meets the requirements at the Checkout Table before the contest begins. See Section 2.5 on the facing page.
As described in Section 6.5.1.3 on page 35, the robot may operate in Sound Activated Mode: it will start when it detects a sound of a specific frequency and amplitude.
The robot’s microphone must have these characteristics:
The Judge will position the Sound Start Device (Appendix A on page 91) approximately 25 mm away from the microphone and will attempt to align it perpendicular to the microphone’s entrance port. Teams may not request any particular orientation or distance.
Figure 2.2 shows a sample Microphone with optional labeling. Note that you must provide a blue background even if the microphone is located on a blue circuit board.
You should verify that your robot responds to the Standard Sound Start Device at the Checkout Table before the contest begins. See Section 2.5 on the preceding page.
Robots using Sound Activation Mode must also have a Start Button as described in Section 2.7.1 on the facing page.
The robot may also have a Power Switch that disconnects the robot’s batteries.
The team may turn the robot on using the Power Switch after placing the robot on the Judge’s table, but the robot must not move as a result.
We recommend that robots be turned on and ready to start before being placed on the table, unless that would cause an unsafe condition. Please discuss your robot’s operation with the Judges if you anticipate a problem.
The team must download any required program or firmware to the robot before it is put on the Judge’s table. The Judge will press only the Start Button or activate the Sound Start Device to start the robot.
The contest arenas will be assembled and available for unscheduled test trials on Saturday.
Due to the limited number of arenas and the large number of robots, waiting lines can become very long.
You should use the practice time to calibrate sensors for the conditions in the gym and to troubleshoot any last minute problems. No team has ever accomplished extensive code development and hardware design on Saturday.
Robots should be built, programmed, and ready to compete on arrival at the contest site. Get busy now!
Some teams bring entire practice arenas along to the competition. You may be able to wheedle your way into those arenas, but that depends entirely on your negotiating skills.
Do not expect any practice time on Sunday morning, although a few arenas may be available for a brief time.
Only one robot is allowed in a practice arena at any one time.
If two robots collide during practice in an arena and one is damaged, then either:
The decisions of contest officials concerning:
are final and cannot be appealed.
Because we do not monitor practice sessions, you are responsible for the safety of your robot at all times.
Power will be distributed as 120 VAC 60 Hz. Your equipment must draw less than 10 A from a single US-standard 15 amp outlet.
You must bring along any voltage or frequency converters required to adapt that power to your needs.
You must bring along sufficient extension cords and outlet strips; you will have access to a single outlet that may be 10 meters from your assigned table in the pit area.
Because the power distribution involves cables laid on the floor, you must assume that power to your devices can be interrupted at any time: people occasionally stumble over the cables and circuit breakers may trip without warning.
Teams must not bring extension cords or external power supplies, such as laptop power bricks, into the arena area. This applies during the Saturday practice sessions as well as the Sunday contests.
Contestants must bring any and all materials, parts, and test equipment that they may need. The Hartford area has very few retail suppliers of electronic and mechanical parts; those suppliers are generally closed during weekends.
The gymnasium is well-lighted, but it is not air-conditioned. Spring weather in Hartford tends to be warm and humid with occasional chilly rain, so plan your wardrobe accordingly.
The Trinity College Firefighting Home Robot Contest (TCFFHRC) advances robot technology and knowledge by using robotics as an educational tool. A winning robot must respond to a fire alarm, discover the blaze, and extinguish it in the shortest possible time.
To accomplish that overall task, the robot must start on a signal, explore a typical family home (the arena), locate a fire (a burning candle), extinguish it, and optionally return to its starting point.
Additional contests held during the TCFFHRC weekend provide different challenges, as described in these rules.
Direct questions and comments about the contest to the Contest Director: Dave Ahlgren david.ahlgren@trincoll.edu.
In order to make the contest accessible to persons of all ages and skill levels the TCFFHRC offers prizes in several Divisions:
Teams or individuals may also demonstrate their robotics knowledge by taking the Robot Olympiad exam (Part V on page 73) and entering the Poster Contest (Part VI on page 77).
Participants who meet the criteria for a particular Division may, at their option, decide to enter their robot in a higher Division, however, they may not enter in a lower Division.
When registering for the contest, each team must specify the robot’s Division. If that Division is full, the robot will be placed on a waiting list.
In order to change Divisions, the team must re-register the robot and pay the appropriate registration fee.
Division entry fees will not be refunded after registration.
No single robot may be entered in two Divisions.
The scoring system emphasizes reliability by grouping robots according to the number of successful trials.
Within each reliability group robots are ranked according to score. To earn a cash award a robot must complete at least two successful trials. Within any contest Division only one prize will be given to any winning robot. However, a robot may win a prize in a contest Division and win one or more special prizes (Cost Effective, etc).
The TCFFHRC awards cash prizes provided by our contest sponsors and non-cash prizes provided by contest supporters. All prizes are described on the Contest Website at http://www.trincoll.edu/events/robot/.
Each team participating in the contest will receive a Certificate of Achievement and one official contest T-shirt.
The World Champion BURP Prize recognizes the best overall performance by an individual or team in the Junior, High School, Senior, or Walking Divisions. We will compute each team’s or individual’s BURP score by weighing
A team or individual must participate in both the Contest and the Olympiad events to be eligible for the BURP award.
The ranking of the robots and teams within their respective Divisions determines their total BURP score. The actual contest scores are not used, only the rankings within the respective Divisions.
This weighting applies to the rankings:
15 robots compete in the Junior Division. This robot wins 4th place.
Score is (4/15) * 0.67 = 0.179
4 teams take part in Olympiad; this team wins 1st place.
Score is (1/4) * 0.33 = 0.083
Total BURP score = 0.179 + 0.083 = 0.262.
45 robots compete in the High School Division. This robot wins 8th place.
Score is (8/45) * 0.67 = 0.119
12 teams take part in Olympiad; this team wins 7th place.
Score is (7/12) * 0.33 = 0.193
Total BURP score = 0.119 + 0.193 = 0.312.
The Junior team has a lower score than the High School team, so its BURP ranking is better.
Once Upon A Time, a creative engineer developed a unique two-legged firefighting robot. Even though the robot was not the fastest in the contest and and had no chance to win first prize, it made its way through the arena and extinguished a candle.
We were so impressed that we created a special award to recognize this engineer’s achievement: The Spirit of the Inventor Award. This award will be given in addition to any other prizes that the robot may win.
To qualify for The Spirit of the Inventor award, the robot must:
The robot need not successfully complete a trial run according to the rules of its Division.
Robotics does not have to be expensive: spending more money does not guarantee success. In fact, some of the very best robots have been some of the least expensive. To award financial efficiency there will be a special prize for the best performing robot built with the lowest amount of money in material cost.
If you put in $50,000 in labor and destroyed $5,000 in parts finally getting it to work, but your final robot has less than $200 in actual parts in it, then it is a good contender for this prize.
It does not matter what you paid for the parts, but only what they are worth. A motor that originally cost $50, but is now for sale in a surplus catalog for $5 is now a $5 motor. However, if you got a $50 motor for free from a friend, then it’s still a $50 motor even though you got it for free. On the other hand, if you destroyed three $50 motors in building the robot, you only have to account for the one motor that is actually on the robot.
Evaluation Method:
Although the contest rules for each Division require robots to fit into a specified maximum volume, there is no minimum size. We challenge teams to build the smallest robot able to successfully complete at least one of its three trials. The robot may compete in any Contest Division.
Size will be determined by measuring the area of the robot’s projection on the arena floor - the smallest rectangle enclosing its chassis and all of the projecting sensors, wires, and appendages.
The size of an Expert Division robot swarm will be the smallest rectangle enclosing all the robots arranged in their starting position. These robots may be stacked if that’s how they will start during their trials.
The Judges will measure all robots competing for this prize.
The arena dimensions and specifications listed below are not exactly what will be encountered at the contest: they are provided as general aids.
The size limits on robots are, however, absolute and will be enforced by the Judges.
The Basic Arena presents a simplified model of a typical house, with high-contrast walls and floors.
The Standard Arena Layout represents a decorated home, a more realistic fire-fighting environment.
The Standard Arena has the same dimensions as the Basic arena. The differences between the Basic Arena and the Standard Arena are listed below.
Once turned on, the robot must be autonomous: self-controlled without any human intervention. Fire-fighting robots are computer controlled, not manually controlled devices.
A robot may bump into or touch the walls of the arena as it travels, but it cannot mark, dislodge, or damage the walls in doing so. The robot must not leave anything behind as it travels through the arena. It must not make any marks on the floor of the arena that aid in navigation as it travels. Any robot that, in the Judge’s opinion, deliberately damages the contest arena (including the walls) will fail that trial. This does not include any accidental marks or scratches made in moving around.
The robot must fit in a Bounding Box with a base 31 x 31 cm square and 27 cm high. If the robot has feelers to sense an object or wall, the feelers will be counted as part of the robot’s total dimensions.
Robots competing in the Walking Division may be up to 46 cm long. The width and height remain as described above.
Robots must not exceed the maximum dimensions at any time, including while extinguishing the candle. This rule prohibits swinging snuffers, extending arms, and other devices that protrude beyond the allowable base or height dimensions while in operation. Team members must demonstrate the maximum extent of any extending devices to the satisfaction of the Judges prior to their first trial.
The robot cannot separate into multiple parts. Robot swarms may compete in the Expert Division (Part III on page 43).
Contestants may add a flag, hat, or other purely decorative, non-functional items to the robot as long as the item has absolutely no effect on the operation of the robot.
Unlike the arena specifications, the robot size limits are not approximate: robots must not exceed the given dimensions.
There are no restrictions on robot weight or materials.
There is no restriction on the type of sensors that may be used as long as they do not violate any of the other rules or regulations. The robot must not extend any sensors beyond the dimensions specified in Section 5.2.2 on the facing page.
Robots using laser-based devices must take measures to prevent eye damage to team members and to observers. The Judges may require the team to remove the laser device from the robot if, in the opinion of the qualification Judges, effective safety measures have not been taken. The robot will be permanently disqualified from competing if the laser cannot be either removed or made safe.
Contestants are not allowed to place any markers, beacons or reflectors on the walls or floors, whether inside or outside of the arena, to aid in the robot’s navigation.
Ambient lighting in the contest room is a mixture of IR, visible, and UV light. During the course of the contest, sunlight may come into the contest room through open outside doors. The sunlight will not shine directly on the arenas, but may be detectable by very sensitive sensors.
During the course of the contest, Judges at other arenas will be lighting candles or lighters. These incidental flames will be above the arena and further away than the candle, but still may be detectable by an undiscriminating sensor. In setting up the arena, contest officials may put their arms into the arena and some very sensitive sensors may mistake that IR emission as the flame.
Many video and still cameras transmit infrared light as part of their automatic focusing systems. Flash units produce bursts of UV that may trigger the popular Hamamatsu UVTron flame sensor. The gymnasium will have many, many cameras at all times: verify that your robot will operate correctly when it’s being photographed.
If a robot uses light sensors to find the candle or detect walls or furniture, the robot designer must prevent unintended UV, visible and IR sources from interfering with its operation. Part of the challenge of this contest is to design a robot that can find the flame and ignore everything else.
AC Power is not available in the arena area.
For obvious reasons of safety and economy, fires will be simulated by small candle flames.
The candle flame will be from 15 cm to 20 cm above the nominal floor level. The candle thickness normally will be between 2 cm and 3 cm. The exact height and size of the flame will change throughout the contest depending upon the condition of candle and its surroundings. The robot is required to find the candle no matter what the size of the flame is at that particular moment.
The candle will be placed at random in one of the rooms in the arena. The candle has an equal chance of being in any of the 4 rooms in each of the robot’s 3 trials. It is possible for the candle to be in the same room on two of the robot’s three trials. If it happens that the candle is placed in the same room for both the 1st and 2nd trials, then the contest officials will make sure that it is a different room for the third and last trial. Thus every robot will have the candle in at least 2 rooms and possibly 3, during its 3 trials.
The candle will not be placed in a hallway, but it might be placed just inside a doorway of a room. The Candle Circle will not touch the doorway line and this means that the front of the robot will be able to move at least 33 cm into the room before it encounters the candle.
The contestants cannot measure or touch the candle before it is used. Violation will result in immediate disqualification of the team and the robot from the competition.
The candle will be mounted on a small wooden base painted semi-gloss yellow. This base prevents the candle from tipping over easily, but a robot can knock the candle over by bumping into it. Judges will give penalty points if that occurs (Section 6.5.4 on page 39).
The robot must, in the opinion of the Judges, have found the candle before it attempts to put it out. For example, the robot cannot just flood the arena with CO2 thereby putting the flame out by accident.
The robot must not use any destructive or dangerous methods to put out the candle.
The robot may extinguish the candle by blowing air or other oxygen-bearing gas. However, this is not a practical method of extinguishing a fire in the real world, so robots that do not use air streams to blow out the candle can operate in Non-Air Extinguisher Mode for an improved score. See Section 6.5.1.6 on page 35 for details.
The robot must come within 30 cm of the candle before it attempts to extinguish the flame. There will be a white 30 cm radius solid circle (or circle segment, if the candle is near a wall) on the floor around the candle, and the candle will be placed in the center of the circle. The robot must have some part of its body over the circle before it extinguishes the candle flame.
Candle Location Mode omits the candle circle and minimum distance requirement. See Section 6.5.1.11 on page 38. The robot need not be within 30 cm of the candle, but must demonstrate that it has detected the candle before extinguishing it. This may be by a distinctive action, an illuminated LED, or other means.
A penalty is given to robots that touch a lit candle with either the robot chassis or a sensor.
Robots may extinguish the flame using air, inert gas, water mist/spray, or mechanical means. The use of powders of any type is not allowed.
A fan is an example of an air-based extinguisher.
Robots may use a single CO2 capsule containing up to 16 grams to extinguish the candle on each trial; larger CO2 containers are prohibited. The Judges will verify that CO2 is the extinguishing material.
Water is the only liquid allowed in this contest. You may not add foaming or gelling agents.
The water tank volume must be no larger than 50 ml. Judges will verify the tank volume.
Water must be applied only as a mist or spray, not a jet.
Any robot that floods the room will fail that trial.
A wet sponge or snuffer.
The size limits described in Section 5.2.2 on page 30 apply to mechanical extinguishers: the robot’s moving parts must not exceed the maximum size at any time.
Carbon dioxide, water mist, and mechanical means qualify for the non-air extinguisher deduction. See Section 6.5.1.6 on page 35.
The robot must perform certain operations during each trial in the arena. This section describes the overall requirements for each Division. Other sections of this document provide further details.
The robot may use any of the available Operating Modes (Section 6.5.1 on page 34) to improve its score for the trial. The robot may use different Modes in different trials, but the team cannot change Modes after a trial begins.
Each successful trial consists of the following sequence of steps.
In general, any robot may operate using any of the Operating Modes available in its Division to improve its score.
Uneven Floor Mode (Section 6.5.1.9 on page 36) is required for robots competing in the High School and Senior Divisions.
Variable Door Locations Mode (Section 6.5.1.10 on page 36) is required for robots competing in the Senior Division and can dramatically improve scores in other Divisions.
See Section 6.5.1 on page 34 for a complete description of the Operating Modes.
Although the scoring system appears complex, it measures differing robot capabilities in different Divisions. The overall scoring flow follows this pattern, with some variations in each Division:
During the trial, the Judges will:
After the trial has completed, the Judges calculate the Operating Score (OS) from those values using this procedure:
Although the “units” of the Operating Score appear to be seconds, they bear little relation to actual wall-clock time.
Scoring rules convert the Operating Score into the Final Score for each trial. The Junior and Walking Divisions share one set of scoring rules; the High School and Senior Divisions share a second set of scoring rules. The Final Score becomes a component of the Total Final Score (TFS) used to rank the robots for prizes and awards.
If the robot extinguishes the candle, then the Final Score for that trial equals the Operating Score. If it did not extinguish the candle, then the robot receives a score of 600 with credit for tasks completed during the unsuccessful trial by subtracting points as described below.
Although a robot with only two successful trials can therefore have a lower Total Final Score than a robot with three successful trials, the ranking described in the next section will award higher prizes to the latter.
TASK.sound = -30
The robot must start properly with sound activation.
TASK.search = -30 x number of rooms searched
Deduct 30 points for each room searched before finding the candle. The maximum reduction is 120 points because the candle must be in the fourth room.
TASK.detect = -30
The robot must correctly signal that it detected the candle by lighting an LED or making an obvious motion.
TASK.position = -30
The robot must stop within 30 cm of the candle without touching it.
The Final Score is equal to the Operating Score: FS = OS.
The Final Score is equal to the Operating Score: FS = OS.
The Walking Division uses the same scoring rules as the Junior Division. See Section 6.2.1 on the previous page.
After all robots within a Division have completed their trials, the Judges compute the Total Final Score (TFS) for each robot by adding all three of its Operating Scores together.
The Trinity Home Firefighting Robot Contest rewards reliable operation by grouping the robots according to the number of successful runs, then according to their Total Final Scores within each group. As a result, a more reliable robot with a worse TFS will outrank a less-reliable robot with a better TFS and be eligible for higher prizes.
The robots in each Division will be divided into four groups based on the number of successful trials: 3, 2, 1, or 0. Within each group the robots will be ranked on the basis of their Total Final Scores. The First, Second, and Third prizes in each Division will be awarded to the three robots with the smallest TFS in the first group. If the first group has fewer than three robots, then the prizes for that Division will extend to the robots with the smallest TFS in the second group, and similarly to the third group.
In all cases, a robot must extinguish the candle in at least two trials to be eligible for a cash award.
These sections explain how the Judges assign values that determine the Operating Score.
A robot’s overall performance depends on its ability to handle real-world situations. The Basic contest arena includes a level floor, high-contrast walls, and no obstructions, but additional operating modes allow you to improve your robot’s score by completing more difficult tasks.
Operating modes act as multipliers to the Actual Time required for the robot to find and extinguish the candle. If no Operating Modes are in effect for a trial, the Actual Time is multiplied by the Standard Mode, which is exactly 1.0.
The team can select different Operating Modes for each of the three trials. Note that the candle and any furniture will be placed in different locations for each trial.
The modes do not apply to an unsuccessful trial. The score for an unsuccessful trial is 600, regardless of any operating modes applied to that trial.
OM.standard = 1.0
The team must inform the Judge of any operating modes for the current trial before the trial begins. In the absence of that notification, the robot will compete in Standard Mode and the Actual Time will be multiplied by 1.0.
Robots tethered by wires to computers, power supplies, or other devices are not permitted, so there is no Tethered Mode.
Robots may communicate through a wireless link, but must operate autonomously. Remote control by a human operator is not permitted!
OM.sound = 0.80Note The Sound Activated Mode factor was 0.95 in previous contests.
The robot begins operation when it detects a sound signal of 3.8 kHz.
The Judges will begin timing the trial when the sound signal begins, not when the robot begins moving. The sound will last 5 seconds and will not be repeated.
The robot must not start until the Judge activates the sound signal. If the robot mistakenly detects ambient noise (even an activation sound from a different arena) and begins to move, then the trial will have begun, but the Sound Activated Mode factor will not apply to the robot’s score.
If the robot does not start in response to the sound signal it will not be given a second chance to use Sound Activated Mode for that trial. The Judge will attempt to activate the robot by pressing its Start Button, but the delay will be included in the robot’s Actual Time for the trial.
Judges will use only Standard Sound Start Devices as described in Appendix A on page 91 during the Contest. Teams should build their own Sound Start Devices and use them during practice, but may not present them to the Judge during the contest.
OM.start = 0.80
The Judge will place the robot in an arbitrary location and orientation within any room that does not have the candle, as determined by the toss of a die.
The robot may be facing a wall or pointed into a corner, but will not be trapped by furniture.
There is no “Home Circle” in Arbitrary Start Mode.
The starting room does not count as a searched room for the Room Factor calculation (Section 6.5.3 on page 38). When the robot leaves the starting room, the next room it encounters is its first searched room.
OM.return = 0.80
The robot must return to its starting location after extinguishing the flame.
In Standard Mode, the robot must return to the Home Circle. It must stop with any part of its chassis within the 30 cm white Home Circle, but need not be in the same position or orientation as when it started the trial.
In Arbitrary Start Mode, the robot must return to the room it started from. It must stop with all parts of its chassis within the starting room, but need not be in the same position or orientation as when it started the trial. See Section 6.5.1.4.
The robot’s Actual Time (AT) recorded for the trial will include only the time required to find and extinguish the candle, not the time for the return trip.
The robot must return its starting location within 2 minutes; if not, then the Return Mode factor is not in effect.
The robot need not retrace its path in returning to the starting location or take the most efficient route, but it must not enter any other rooms along the way.
OM.extinguisher = 0.75
The robot must extinguish the candle using inert gas, water, or mechanical means. See Section 5.3.1.1 on page 32
Robots that use an air stream of any kind do not operate in Extinguisher Mode.
OM.furniture = 0.75
Every room will have one or more pieces of furniture. This includes the room where the robot starts in Arbitrary Start Mode.
Furniture consists of semi-gloss yellow cylinders 11 cm in diameter, 30 cm high, and weighing more than 1 kg.
Furniture will always be placed to allow at least one path to the candle that is at least 31 cm wide. The furniture will not block the doorway and a maximum-size robot will be able to come into a room at least halfway before it encounters furniture. Furniture may block the robot’s view of the candle, so it must move to different locations to see the candle and plan a path to reach it.
The robot may have to go around the furniture to extinguish the candle or exit from the room. It may touch the furniture, but it cannot push it out of the way. Robots that push the furniture away lose the Furniture Mode deduction for that trial.
OM.coattree = 0.80.
A small coat tree, shown with dimensions in Figure 6.1, may be placed in any room or hallway. Clothing items with various cloth textures and colors will hang on the coat tree.
The robot must not move the Coat Tree. Robots that move the Coat Tree lose the Coat Tree Mode deduction for that trial.
The Coat Tree will not block the robot’s passage through a hallway.
A Coat Tree within a room will follow the placement guidelines in Furniture Mode (Section 6.5.1.7 on the preceding page).
The coat tree has a hardwood base and an upright made from a wooden dowel 1.6 cm in diameter. Four 3.5 cm clothes pegs are inserted into the dowel at a 45 degree angle 3.5 cm from the top. The tree is 25 cm high.
OM.floor = 0.80
The robot must operate in an arena with ramps placed in hallways to defeat dead-reckoning navigation.
The ramps have a 15 degree maximum slope and a 5 cm maximum height. The ramps are tapered with discontinuities less than 5 mm. The ramps have the same flat-black paint as the floor.
More than one ramp may be present in the arena for any trial and the exact placement of ramps will be unknown to the robot before the start of any trial. The ramps will not be placed in the hallway directly outside of a doorway, although one could be placed adjacent to a doorway. The ramps will remain in place during the return trip portion of the trial.
The number and location of the ramps will be changed from trial to trial.
OM.variabledoor = 0.45
This option presents uncertainty about the locations of the Room 1 and Room 4 doors.
At the start of a trial the arena Judge will determine the door locations by tossing a die or using a computer-assisted method. Therefore, the robot may encounter a different door location on each trial.
Figures 6.2, 6.3, 6.4, and 6.5 show all possible door locations. Shaded areas mark possible rug positions.
OM.candle = 0.75
This option challenges robots to find candles without a candle circle. The Judge will place the candle at a randomly chosen location within a room for each trial.
The candle may be in any location within the room that does not block the doorway. A maximum-size robot can enter the room at least halfway before encountering the candle and there will be at least a 31-cm wide path around the candle.
The candle won’t be directly adjacent to a wall, to reduce the chance of damaging the wall by overheating. There is no specification for the exact distance from the wall.
There are no other restrictions on the candle location in this Mode.
The Fire rules in Section 5.3 on page 31 will be followed except that:
The Furniture Mode rules in Section 6.5.1.7 on page 35 also apply in Candle Location Operating Mode. In particular:
If the robot extinguishes the flame the Actual Time is the number of seconds elapsed from robot activation to flame disappearance. The maximum Actual Time for such a successful trial is AT = 300. If the robot does not extinguish the flame within the limits set below, the Judge will terminate the unsuccessful trial and assign AT = 600.
The maximum time allowed for a robot to find the candle is 5 minutes, after which the Judge will stop the trial and assign AT = 600.
A robot operating in Return Trip Mode must return to the Home Circle within 2 minutes after extinguishing the candle, after which the Judge will stop the trial. The AT equals the time required to extinguish the candle.
If a robot gets stuck in a loop and performs the same (or a similar) movement 5 times in a row without progress, the Judge will stop the trial and assign AT = 600.
Any time the robot does not move at all for 30 seconds, the Judge will stop the trial and assign AT = 600.
A robot that fails at both of its first two trials will not receive a third trial.
The Room Factor (RF) adjusts the elapsed time based on the number of rooms searched. The more rooms a robot searches before it finds the candle, the lower the Room Factor for that trial.
When the candle is in:
It does not matter in which order the robot searches the rooms. The only thing that matters is how many rooms the robot has searched before it finds the candle.
When the robot searches the room with the candle, whether or not the robot extinguishes it, the Judge records the Room Factor for that trial. The room factor will not change regardless of how many more rooms the robot searches.
Because some robots can detect the candle by looking in the doorway without entering the room to search it, when the robot passes a doorway for the first time the Judge will count that room as searched. If the robot has already searched a room and then goes past the doorway again on its way to a different room, that room will not be counted twice.
Penalty Points (PP) will be added to the Actual Time (AT) of any robot that exhibits the behaviors described in this section.
PP.button = 35
The robot must have a Start Button with the characteristics described in Section 2.7.1 on page 18, even if the robot operates in Sound Activated Mode.
The Judges will determine whether the Start Button conforms with the specifications when the team presents the robot for each trial. If it does not, the Judges will add penalty points to the score for that trial.
Teams should verify that their robot’s Start Button meets the specifications at the Checkout Table described in Section 2.5 on page 18 during the practice time before the contest begins.
PP.candle = 50
Any robot that touches the candle or its base, either deliberately or accidentally, while the candle is lit will have 50 penalty points added to its Actual Time score each time the candle is hit.
There is no penalty for a touch that occurs as part of the actual extinguishing process, i.e., smothering the flame with a wet sponge, or after the candle is extinguished.
Touching refers only to any part of the robot’s body, including feelers or probes, and does not include the water, air or other material that the robot might use to extinguish the candle.
Although there is no penalty for touching or knocking the candle over after the robot has extinguished the candle, we strongly recommend that your robot avoid doing that. The Judges may not agree with your opinion of whether the candle was extinguished before it began falling.
PP.slide = (contact cm) / 2
Any robot that slides along a wall will have 1 point added to its Actual Time score for each 2 cm of wall it touches.
A robot may still touch a wall to orient itself, as long as the contact is not sliding.
There is no penalty for touching or sliding along the wall on the return trip to the Home Circle.
See the Note in Section 5.2.1 on page 30 regarding “Navigation by Crashing”.
These examples illustrate how to calculate the Total Final Score under specific conditions for each Division.
Any disagreement between these examples and the rules given above will be decided by reference to the rules!
The robot uses Sound and Return modes in its first attempt, takes 1 minute and 23 seconds to extinguish the candle in the 2nd room and slides along the wall a total of 42 cm. The robot puts out the flame with a fan.
The robot uses Sound and Return modes in its second attempt, but fails to return to the home position; OM.return is not in effect. The robot takes 1 minute and 41 seconds to blow out the candle in the fourth room searched. It accidentally bumped the candle one time.
The robot navigates to two rooms, indicates that it sees the candle, but does not extinguish the candle or come within 30 cm of the candle. The robot starts with an audio signal. There are no penalties.
Total Final Score: TFS = 56.576 + 42.28 + 510 = 608.856
Ranking: two successful trials = second group.
The Walking Division will be scored in the same manner as the Junior Division. See Section 6.6.1 on page 39.
The TCFFHRC Expert Division extends the contest presented in the Senior Division to include more realistic situations, configurations, and challenges.
Three challenges will have a major effect on robot design:
Unless otherwise noted in this Part, all of the rules and requirements in Part I on page 9 apply to the Expert Division.
The Expert Division contest takes place in a slightly modified TCFFHRC Standard arena. See Section 5.1.2 on page 29 for the overall configuration. This section describes the differences from that configuration.
The doorway locations to two of the four rooms will change for each trial: the robot will encounter a different door configuration on each of its three trial runs.
Therefore, the robot must run in Variable Door Locations Mode, as described in Section 6.5.1.10 on page 36.
Every room will contain at least one and not more than three Furniture items, consisting of scale-sized furniture normally found in a house: Chairs, Tables, Couches, Ottomans, and Beds, as shown in Figures 7.1 on the following page through 7.5 on the next page.
The furniture may block a view of the fire from the doorway and may block a robot’s path into or through the room.
Furniture will be positioned with one side against a wall and will not be an island in the room. Items such as the Ottoman and Small Chairs will be positioned no more than 2 cm from the nearest furniture item.
A Furniture item may be located directly below a wall sconce, blocking the robot’s approach.
A Furniture item may block the robot’s approach to the Wastebasket from either (or both) sides.
The robot may move the Furniture without penalty, although it must not touch or move the flame container before extinguishing the flame.
The following pictures provide detailed views of the furniture.
Figure 7.6 shows the Chair, which is 15 cm long, 17.5 cm wide, and 17 cm high.
Figure 7.7 on page 49 shows the Chair and Ottoman from above. The Ottoman is 12 cm square and 8 cm high.
Figure 7.8 on page 49 shows the Small Chair. The seat is 18 cm square and the back is 18 cm high.
Figure 7.9 on page 49 shows the Couch (also known as the Love Seat: a small, cozy couch), which is 15 cm long, 31 cm wide, and 17 cm high.
Figure 7.10 on page 49 shows the Table. The top is 18 cm square and 15 cm high.
Figure 7.11 on page 49 shows the Bed. The Bed is about 32 cm long and 21.5 cm wide. The headboard is 21 cm high and the footboard is 12.5 cm high. The cloth color and material will vary from that shown in the Figure.
The arena will have a Coat Tree as described in Section 6.5.1.8 on page 36.
The Coat Tree will be at least 30 cm away from a Wall Sconce (Section 7.2.4.2 on page 51).
The robot must not move the Coat Tree while traversing the hallway or extinguishing an adjacent Wall Sconce fire. Unlike the other Divisions, if an Expert Division robot moves the Coat Tree, the robot fails that trial.
The Expert Division simulates a major house fire by using multiple fires in different locations. The robot will encounter fires in hallways, as well as inside rooms, and must extinguish all the fires in the order it encounters them to have a successful trial.
Because the robot will not know the number or location of the fires before starting the trial, it must search every room and hallway to verify that all fires have been discovered and extinguished. Therefore, the scoring formula does not include a Room Factor.
The robot need not extinguish the closest fire before proceeding in a different direction. For example, upon entering the middle of a hallway with fires at each end, the robot may choose either fire. However, the robot must not pass by a fire without extinguishing it: if the robot passes a doorway to a room containing a fire, it must extinguish the fire in the room before continuing along the hallway.
If a robot passes by the doorway to a room containing a fire or passes a fire in a hallway without extinguishing it, the Judges will record that the robot has missed a fire. This applies even if the robot returns to the missed fire and successfully extinguishes it.
A robot swarm may dispatch several “seeker” robots to search many rooms simultaneously. However, any “extinguisher” robot capable of putting out a fire must do so in the order it encounters the flames.
For example, the seekers may search all rooms and hallways for fires so that the extinguisher robot can compute an optimal path before it starts moving. The extinguisher robot must not pass a room or hallway fire without extinguishing it.
The robot may extinguish a fire from any distance; there is no requirement to be within 30 cm of the flame. The other rules in Section 5.3.1 on page 31 remain in effect; we commend your attention to the Water Rifle Exception ( 3 on page 32).
Fires may occur in any part of the arena, including the hallways.
Hallway fires will be contained in wall Sconces (Section 7.2.4.2 on the facing page).
Room fires may be contained in either Sconces or Wastebaskets (Section 7.2.4.1 on the next page).
The arena will contain two, three, or four fires for each trial, with at least one fire in a hallway and one in a room. The third and fourth fires, if present, may be in any locations.
The robot will encounter any possible fire configuration only once in its three trials.
Robots may be able to see multiple fires from a single location. Examples:
The fires will be flames produced by small candles within containers.
The candles are commonly known as “tea candles”, as shown in Figure 7.12 on the facing page. For more information, see http://en.wikipedia.org/wiki/Tealight.
The aluminum container is approximately 4 cm in diameter and 1.5 cm tall. The height of the flame will vary depending on the amount of wax remaining in the container and the length of the wick.
A candle in a small container simulates a wastebasket fire, as shown in Figure 7.13.
The ceramic container is approximately cubical: 7 to 8 cm on each side.
Although some part of the flame will be visible above the container’s rim, Furniture may block the robot’s view of the flame from some parts of the room. In addition, Figure 7.14 shows that the container may obscure the flame from sensors located below the container’s upper rim.
A candle in a wall Sconce simulates a fire in a window curtain or wall hanging. Figure 7.15 on the next page shows a sample Sconce containing a Candle in a low position.
The Sconces have shallow candle holders, as shown in Figure 7.13, with the candle containers secured to the brackets to prevent spilling.
The flame will be between 3 cm and 200 cm below the top of the arena wall. Section 5.1.1 on page 29 defines the arena wall height, which may vary from arena to arena.
Figure 7.16 on the following page shows a sample Sconce holding a Candle in a high position.
with_small_candle_high-2011-09-07_2.jpg)
Although at least part of the flame will be above the edge of the sconce, a robot with a flame sensor far below the top of the Sconce may not be able to see the flame. See Figure 7.14 for an illustration of the geometry with a different container.
Figures 7.17 on the next page and 7.18 on page 53 show details of the Sconce construction. The platform supporting the candle container can be inserted into any of the holes to adjust the flame height.
Because the Expert Division arena may have multiple fires, a team may deploy multiple robots. There is no limit to the number of robots, subject to the restrictions in this section, and a swarm of small robots is perfectly acceptable. The robots may be identical or different, however, see Section 7.2 on page 50 for the requirements imposed on different robots while extinguishing fires.
All of the robots must fit together within the maximum dimensions described in Section 5.2.2 on page 30. The robots may be arranged in any manner within that Bounding Box, including being stacked, but must meet these requirements:
Robots must use these Operating Modes. There are no optional Operating Modes.
The Judges will place all of the robots in a single location. The team may specify, in writing, the position and orientation of each robot within the starting array (which must fit within the Bounding Box), but not the position or orientation of the array with respect to the arena walls or features. If the robots do not all face in the same direction, some of them may be aimed directly at a wall.
All robots in a swarm must return to the starting location, but need not park in their starting arrangement.
The water tank volume of all robots may be distributed in any manner, but must not exceed 50 ml total. For example:
Furniture may be moved without incurring a penalty.
The Coat Tree must not be moved.
The Expert Division Operating Modes do not have Mode Factors, because the robots must use all of the Modes. As a result, Expert Division scores will be numerically equal to the total elapsed time for the trials.
The Expert Division scoring procedure follows this general outline:
Most of the Operating Modes and Penalties are identical to those used in the Senior Division, with exceptions as noted below.
During each trial, the Judges will:
After the trial has completed, the Judges calculate the Operating Score (OS) from those values using this procedure:
Scoring rules convert the Operating Score into the Final Score for each trial, which then becomes a component of the Total Final Score used to rank the robots for prizes and awards.
If the robot extinguishes all the flames, then the Final Score for that trial equals the Operating Score. If it did not extinguish all the flames or otherwise fails the trial, then the robot receives a score of 600 with credit for tasks completed during the unsuccessful trial by subtracting points as described below.
Although a robot with only two successful trials can therefore have a lower Total Final Score than a robot with three successful trials, the ranking described in the next section will award higher prizes to the latter.
TASK.flames = -30 x 4 x (number extinguished / total flames)
The Task Count is weighted to compensate for the variable number of flames.
TASK.search = -30 x number of rooms searched
The robot must search all four rooms, even if it has already found all four possible flames.
Task.return = -30 if the robot returns to the starting position
All robots in a swarm must return, but they need not be arranged in their starting positions.
After all robots have completed their trials, the Judges compute the Total Final Score (TFS) for each robot by adding all three of its Operating Scores together.
Expert Division Ranking follows the same algorithm as for the other Divisions. See Section 6.4 on page 34 for details.
The Expert Division uses the same scoring components as the other Divisions, as described in Section 6.5 on page 34. However, the Expert Division has more stringent requirements for some components and does not apply Mode Factors. This section describes those differences.
See Section 7.4 on page 53 for the list of mandatory Operating Modes in the Expert Division.
Expert Division scoring does not use Mode Factors, because all Operating Modes are required.
The Actual Time is measured as described in Section 6.5.2 on page 38, with the AT equal to the time until the last flame is extinguished.
Robot swarms have additional requirements:
The Room Factor does not apply to the Expert Division, because the robot must search all four rooms on each trial.
Penalty Points do not apply to the Expert Division. Those infractions that would trigger Penalty Points in the Senior Division cause the robot to fail the trial.
The robot extinguishes all three candles in 2 minutes and 10 seconds.
The robot extinguishes all four candles in 1 minute and 55 seconds.
The robot extinguishes two of three candles, searches all four rooms, bumps the Coat Tree in passing, and returns to its starting location. The Judge assigns AT = 600.
Total Final Score = 130 + 115 + 370 = 615
Ranking: two successful trials = second group.
The RoboWaiter Contest challenges teams to create a robot that can retrieve a plate of food and transport it to a table in a reliable and efficient manner. The arena simulates a home kitchen with the usual fixtures and a pair of dolls representing humans served by the robot.
The 2012 Assistive Robotics Contest (a.k.a “RoboWaiter”) was founded with support from the Connecticut Council on Developmental Abilities.
The competition presents a situation where Grandpa, a person with a disability, wants a plate of food from the Refrigerator. He sits at the kitchen table in his wheelchair. The arena represents a household kitchen, with Grandma, a second chair, a sink, and the refrigerator.
In the Standard Division, a simple shelf represents the refrigerator. The Advanced Division arena refrigerator consists of an enclosed box with an automatic door and two shelves, one of which will contain the food.
When directed by a signal from the Judge, the robot will move to the refrigerator, pick up the plate, and place it on the Table, while avoiding obstacles within the kitchen. Optional tasks include returning to the starting point and moving the plate from the Table to the Sink.
As in all Trinity Robotic Contests, the robot’s action must be fully autonomous.
The Judge will measure and record the time from the start signal until the robot places the plate on the table. Various Operating Modes reward successful completion of more challenging tasks by improving the overall score.
The RoboWaiter Contest has two Divisions.
Standard Division presents a simplified challenge:
Advanced Division robots must use precision navigation and accurate timing:
Appendix B on page 93 presents details of the refrigerator hardware. The Contest Website at http://www.trincoll.edu/events/robot/ may have additional details.
RoboWaiter is open to any team registered in the TCFFHRC. To register for RoboWaiter, check the box on the registration form. Teams may enter kit or unique robots.
A team may enter a robot into the RoboWaiter contest without entering a robot in the TCFFHRC.
See Section 1.3 on page 13 for registration and fee information.
Cash prizes for first, second, and third place will be awarded, plus a special prize for the most successful walking robot in either Division of the RoboWaiter competition.
The competition takes place in a square arena that simulates a kitchen. The arena is 2.5 m on each side, with a black floor and white walls that are 30 cm high.
A Home Circle marks the robot’s starting point for the trial. The circle is white and 30 cm diameter. It will not be secured to the arena floor. The Home Circle position will vary for robots competing in Arbitrary Starting Location Mode (Section 11.2).
Figure 10.1 shows the arrangement of the Standard Division arena.
The Chair is always present. It will be located as shown in Figure 10.1, with its back against the wall, at the midpoint of the wall.
The Grandma doll is optional. If the Grandma option is selected, it will be located and oriented at the positions shown in Figure 10.1.
Figure 10.2 shows the arrangement of the Advanced Division arena.
The Chair is required and will be located as shown in Figure 10.2, no more than 75 cm from the wall. It may be oriented at any angle.
The Grandma doll is required and will be located at a random position along the line shown in Figure 10.2, at most 75 cm (to the doll center) from the wall. its exact location and orientation along that line will vary for each trial run. It will not block access to the refrigerator or Sensor. As mentioned in in Section 10, if the robot touches the Grandma doll, the robot fails the trial.
Dolls similar to these will be used in the arena. For historic reasons, we refer to the doll at the table as Grandpa and the doll standing in the arena as Grandma, but your robot should expect similar dolls of either gender at either location.
Because robots must operate safely around humans, a robot that touches any part of a doll will incur severe penalties.
The Grandpa doll and wheelchair (Figure 10.3) will be positioned at the table.
The Grandma doll (Figure 10.4) is optional in the Standard Division and required in the Advanced Division.
Grandma will be positioned on the floor in the arena as described in Sections ?? on page ?? and 11.1.2 on page 69.
The plate is located on a shelf as described in the Standard and Advanced sections below. Each shelf will have one plate.
The plate is round: 10 ± 0.3 cm in diameter. It is a pet-food can cover, Curtis Wagner Plastics Corp. Item #PF-4200 (http://hometownusastores.com/product_info.php?products_id=1814). See Figure 10.5.
A steel washer glued to the base of the plate adds weight. Also fixed to the bottom of the plate are four plastic feet, which help prevent slippage of the plate on the shelf. The total weight of the plate, including the steel washer and the plastic feet is 50 grams. See Figure 10.6 on the next page.
The shelf supporting the plate is 40 cm deep (front-to-back) and 45 cm wide (left-to-right). The top of the shelf is 20 cm to 24 cm above the floor. The shelf height will change from trial to trial, so the robot must cope with an unknown shelf height! See Section 2.3 on page 17 regarding dimensional limits.
There are three bright red light-emitting diodes fixed to the edge of the shelf, separated by 2.0 ± 0.1 cm center-to-center (Figure B). The mid-point of the plate’s edge is aligned with the middle LED. Figures 10.7 on the next page and 10.8 on the facing page show the LED and plate arrangement.
The table is 70 cm wide (left-to-right) and 50 cm deep (front-to-back), with one bright red LED (Section 10.1.5.1 on the preceding page) at the center of each visible side. The top of the table is 20 to 24 cm above the floor. See Figure 10.9.
The sink serves as an obstacle and has the same footprint as the table. The sink is 25 cm high. A single blue LED centered on the front edge marks the center of the sink bowl.
See Figure 10.10 on the next page.
The blue LED is available from Mouser: 941-C503BBCNCV0Z0462 or Cree C503B-BCN-CV0Z0462.
The chair has a footprint of 20 cm x 20 cm. See Figure 10.11 on the following page.
Figures 10.1 on page 63 and 10.2 on page 63 show the possible chair positions.
The RoboWaiter Advanced Division arena has a simulated refrigerator in place of the Shelf.
The refrigerator has two shelves in an enclosed box, with a door that opens and closes under the robot’s control.
This section describes the refrigerator’s physical dimensions and characteristics. The next section describes the sensor that triggers the door operations.
The refrigerator exterior is 20-25 cm deep (front-to-back) x 45 cm wide (left-to-right) x 42 cm tall.
The refrigerator shelves are 16-21 cm deep (front-to-back) x 42 cm wide (left-to-right).
A continuous, modulated infra-red LED beacon, aimed perpendicular to the face of the door, is located within 1 cm of the center of the refrigerator door. Figure 10.12 gives the dimensions and Figure 10.13 shows the beacon in visible light.
The beacon consists of five IR emitters and one visible emitter on a small circuit board taken from a 6-LED flashlight. Five of the six LEDs were replaced by IR emitters; the remaining visible LED indicates that the device is working.
The beacon emits approximately 300 mW of 880 nm IR with a beam width of approximately 30 degrees. Designers should not assume a uniform or Gaussian intensity distribution within the beam.
The driver circuitry modulates the beacon at 8.0 kHz ± 10%.
A simple IR phototransistor mounted in a flashlight reflector readily detects the beacon from a distance of more than one meter.
See Appendix B.1 on page 93 for details of the beacon design.
Robots may use wall-following techniques on the open door while navigating into and out of the refrigerator, but they must not touch the door while doing so.
Each shelf has three LEDs as described in Section 10.1.5.
Figure 10.14 shows the front view of the Refrigerator interior.
Figures 10.15 through 10.17 on the next page show top views of the Refrigerator with the Door in various positions.
A plate will be located on each shelf, aligned as described in Section 10 on page 63, Item 10.1.5.
The robot must fetch the correct plate as part of a successful trial: if the robot takes the plate from the wrong shelf, the robot has failed that trial.
The robot may touch the shelf (but not the door!) while aligning itself to the plate’s position. However, mechanical lever-action switches may not trip reliably on contact with the shelf, particularly at nearly perpendicular approach angles. If your navigation algorithms depend on switch closures, test your mechanical linkages very carefully under worst-case conditions, because that’s what your robot will encounter at Trinity!
Advanced Division robots must open and close the refrigerator door by triggering a sensor module embedded in the floor directly in front of the refrigerator. The sensor module lies in the center area of the refrigerator’s door beacon pattern, 65 cm from the outside surface of the closed door.
The sensor module contains three bright white visible LEDs that shine directly up from the floor and a Sharp GP2D120 IR proximity sensor. The LEDs are in a line 3 cm from the proximity sensor, parallel to the front of the refrigerator. The entire sensor module is embedded in the floor and will not impede robot motion.
Figure 10.18 on the next page shows the LED and GP2D120 arrangement within the sensor module. The refrigerator is located 65 cm from the upper edge of the rectangle in the figure.
See Appendix B.2 on page 93 for details of the sensor construction
The proximity sensor is connected to the refrigerator’s embedded microcontroller. When the microcontroller first senses the presence of a robot, it will open the refrigerator door. When it senses the robot again, it will close the door. Therefore the sensor acts a toggle switch that controls the door’s opening and closing.
A large robot or one that holds the plate in front of its main chassis may trigger the sensor before it has cleared the path of the closing door. The robot must avoid contact with the door, because the robot will fail the trial if the door touches it while closing.
The door will begin opening or closing within one second of the time the sensor detects the robot.
The door will open or close completely within five seconds from the start of motion.
The robot must move completely off the sensor module for at least 5 seconds while extracting the plate.
The refrigerator door will operate only one time during a trial. The robot must ensure that it does not trigger the door-opening sequence before it is ready to extract the plate.
The robot must fit into a cube measuring 30 cm on a side at the start of the trial.
Although the robot may deploy grippers beyond the starting envelope while transporting the plate, the “no contact” rules apply to all parts of the robot.
The robot need not retract its grippers after placing the plate on the table, but that would be a nice touch.
All robots in the Standard Division must have a Start Button as described in Section 2.7.1 on page 18.
Optionally, the robot may start with the sound signal used to mark the lower shelf in the Advanced Division, as described in Section 10.2.2.2. However, there is no Sound Activated score deduction.
Sound activation is required for robots competing in the Advanced Division. A Start Button is not required for robots in the Advanced Division, because (unlike in the Firefighting competition) if the robot fails to start in response to the Standard Sound Start Device, it will have failed the trial.
The sound frequency indicates which plate the robot must retrieve from the refrigerator:
The starting signal will be a Standard Sound Start Device as used in the TCFFHRC (Section 6.5.1.3 on page 35), with an additional audio output at a different frequency to specify one of two plate locations. See Appendix A on page 91 for details of the Standard Device.
The general rules described in Chapter 2 on page 17 apply unless otherwise noted below.
This section describes the overall procedure of a RoboWaiter contest trial.
Teams may chose optional Operating Modes to modify the challenge. See Section 11.2 on the following page for details. The Operating Mode multipliers apply to the total time required to fetch the plate and place it on the table.
These rules apply in addition to the Standard Division rules in Section 11.1.1.
The robot may touch either refrigerator shelf, perhaps to align itself with the plate, without penalty. However, it must not touch the door at any time.
All robots in the RoboWaiter contest may select any or all of these optional Operating Modes to improve their time scores. Each trial may use a different combination of Operating Modes.
When completed successfully the following options, available on each of the three trials, will result in reduced time scores, using the indicated multiplication factor (MF) for each of the listed premiums.
The plate will contain a simulated food item: meat, potato, etc. Multiplier will be earned by delivering the plate to the table without dropping or spilling the food.
The food will be actual food, such as cereal or pasta, that does not stick to the plate. The food will not add any significant weight to the plate.
The Judge will choose the robot’s starting location and orientation at random and place the Home Circle at that position. This arbitrary starting location will not be physically closer to the plate (on the shelf or in the refrigerator) than the standard starting location. See Figure 11.1 for possible locations.
The Grandma doll is optional in the Standard Division. If this option is selected, the Grandma doll will be located as shown in Figure 10.1 on page 63. If the robot touches the Grandma doll, the Grandma Operating Mode factor will not apply to that trial.
The Grandma doll is always present in the Advanced Division, the Grandma Operating Mode does not apply. If the robot touches the Grandma doll, it fails that trial.
The robot must return to the Home Circle position where it started the trial. The robot need not be in the same orientation as when it started the trial.
The robot’s Actual Time (AT) recorded for the trial will be the time required to transfer the plate to the table, not including the return trip. However, the robot must return its starting location within 2 minutes; if not, then the Return Mode factor is not in effect for that trial.
After the robot has returned to the Home Circle and ceased all motion, the Judge will restart the robot using the 3.8 kHz Standard Starting Device tone. The robot will return to the table, pick up the plate, transfer it to the sink, then return to the Home Circle again.
The plate may be dumped into the sink in any orientation, but it must remain either on the top surface of the sink or within the bowl.
The robot must complete the Clean Up operation and cease all motion within 4 minutes of the Start Signal. That time is not added to the total time: the Clean Up Mode factor is applied to the original time required to place the plate on the table. In effect, by completing the Clean Up operation, the robot receives a significant scoring advantage.
However, if the robot fails to complete the Clean Up operation, it will fail the entire trial.
Each robot will compete in three trials, with the Judges recording the time required to complete the trial. Section 11.2 on the preceding page describes the Operating Modes in detail. Section 12.0.1 provides examples of scoring calculations.
Robots that have successfully completed at least one trial will be eligible for First, Second, and Third prizes, with accompanying cash awards. Robots that have not successfully completed at least one trial will not be eligible for prizes or cash awards.
Successful robots will be divided into three groups, based on the number of successful runs, to ensure that the most reliable robots receive awards. The ranking within each group will be based on the robot’s final score for all three runs. The groups are:
Winners will be taken starting with the highest-ranking robots in the Most Reliable group, then continuing with the Moderately and Least Reliable groups, until the three winners have been identified.
Examples:
You can help by creating scoring examples for various situations, using the rules as a basis, to help verify your understanding of the rules. Please send them to the Editor (Ed Nisley ed.nisley@ieee.org).
Robot starts at home position, finds plate and delivers it to the table. No food on plate. No mode options completed.
Measured Actual Time AT = 78 s
Score: Success = 1; Time = AT = 78 s
Robot starts at home position, finds plate with food on it, delivers plate to table, returns to start.
Measured Actual Time AT = 56 s
Score: Success = 1; Time = AT * OM.food * OM.return = 56 * 0.8 * 0.8 = 35.84 s
Robot starts at arbitrary position, delivers plate with food to table, returns to start, and completes cleanup option.
Measured Actual Time AT = 109 s
Score: Success = 1; Time = AT * OM.start*OM.food * OM.return * OM.cleanup = 109 * 0.85 * 0.8 * 0.8 * 0.7 = 41.507 s
Success = 3; Time = 78 + 44.8 + 41.507 s = 182.1 s
The robot is placed in the Most Reliable group with three successful trials. Its ranking will be determined by comparing overall time scores within that group.
The TCFFHRC Olympiad exam consists of about ten questions, each presenting a real problem that might arise during robot design projects. Each question requires a solution based on theoretical background and practical experience.
The exam takes 50 minutes.
The Olympiad is open to any registered team or individual, and prizes will be awarded to teams and individuals in Junior, High School, and Senior Divisions.
Check www.trincoll.edu/events/robot for the 2012 Olympiad schedule.
Questions about the Olympiad may be directed to:
Contributed by David Pietrocola, Science Education Analyst, National Science Foundation
The ability to effectively communicate technical ideas and designs is an increasingly important skill for engineers and scientists. The 2012 TCFFHRC technical presentation competition aims to encourage the development of such communication skills. We encourage teams to summarize and convey their efforts by designing and delivering a presentation that explains the design and functionality of the robot. The competition is entirely voluntary.
Teams may select from two tracks:
The Poster Format track involves designing a poster following established scientific poster templates as described below. The New Media track encourages teams to use video, audio, and other interactive media formats to present their robot design.
Cash prizes will be awarded for the best submission from each track.
Both tracks will be judged based on the same criteria: 40% content, 30% visuals, 30% presentation.
The poster presents the design of the team’s firefighting or assistive robot. Following traditional scientific poster templates, posters are encouraged to include the following sections and components:
Teams will register for this track as part of the web-based contest registration process. Poster stands will be provided to those who register for the poster session.
Poster size limits:
On Saturday, March 31, two members of each team will give a 5-minute, focused oral presentation to the judges. Teams must make their presentation in front of their poster at their assigned time, which will be available at check-in. Presentation of the physical robot to the judges will not be permitted.
All posters and presenters must use English. However, teams for whom English is a second language may request the assistance of an interpreter during the presentation by checking the appropriate box on the registration form and indicating the language.
The posters will not be evaluated as part of the contest and do not contribute to the robot’s Best Unified Robot Performance (BURP) score.
This track encourages the use of video, animation and other computer-aided tools to present the design of a team’s firefighting or assistive robot. The content of the video or interactive element should be similar to what was described for the poster track (see above), with a maximum duration of five minutes.
In lieu of an on-site presentation during the contest, submissions must be narrated by team members, though they are not required to be “on screen” the entire time.
Submissions must be hosted on streaming websites such as YouTube or Vimeo.
Teams will register for this track as part of the web-based registration process. A link to the video submission must be submitted to contest organizers no later than Saturday, March 24.
All audio and text used in video submissions must be in English.
Trinity College’s Fire-Fighting Home Robot Contest rules are published on the Contest Website at http://www.trincoll.edu/events/robot/.
We invite you to use these rules without charge for the limited purpose of use as the basis for a non-profit educational project or to organize your own non-profit firefighting robot contest. You acknowledge and agree by your use of these rules, whether for an official regional contest or an unofficial contest, that Trinity College assumes no responsibility or liability for such use of the contest rules by you or any third parties. These rules are provided “as is” without any warranty of any kind.
If you plan to use the Trinity rules, we request that you send a 50-100 word description of your activity to the contest Director via email.
Note that your use of the Trinity rules does not automatically qualify your robot to participate in the official Trinity College Fire-Fighting Home Robot Contest (“TCFFHRC”) to be held at Trinity College.
Official regional contests are public events based on the Trinity rules found on the Contest Website at http://www.trincoll.edu/events/robot/. The characteristics of official regional contests and Trinity’s relationship to them are listed below.
In order to hold an official regional contest, the contest should meet these requirements:
In order to become an official regional contest and to obtain the benefits listed above, please send the contest director an email message indicating your interest and confirming your agreement to the requirements described above. In turn you will be sent an application form that asks such information as name and date of event, expected participation, contest Divisions that you wish to offer, and names of sponsors.
When planning your event please note that normally regional contests are held within eight weeks prior to the official Trinity College Fire-Fighting Home Robot Contest to be held at Trinity College.
Requests for new regional contests should be sent to the Director at least six months before the next Trinity contest.
Judges will use only the Standard Sound Start Device during the contest. Teams may not bring their own devices into the arena area during trials.
The Sound Start Device emits two selectable tones: 3.8 kHz and 2.5 kHz.
The specified sound modules produce approximately 90 dB SPL at 1 foot. The SPL will be higher at the microphone, due to the closer distance, but there is no specification for the actual intensity.
The selected tone will sound for at least 5 seconds after the Judge presses the Tone button.
The robot must start with the Sound Start Device approximately 25 mm from the robot’s microphone. The Device has a 25 mm rod indicating this distance; the rod will not touch the robot.
Figure A.1 shows a Standard Sound Start Device.
Figure A.2 shows the schematic diagram of the circuitry inside the Sound Start Device.
How it’s built, along with the 25 mm rod requirement
Figures B.1 and B.2 show the LED array and driver circuit board.
Figure B.3 on the next page gives the beacon driver schematic and parts list.
Figure B.4 shows the sensor module installed in the black-painted arena floor.
Figure B.5 on the preceding page shows the sensor schematic and parts list.
