ASEE 2019 Competition: Rescue Robots

The ASEE 2019 robotics competition will be held in Tampa Florida, each year the ‘game’ is designed based on the location of the competition. Florida endures many hurricanes each year so this year the competition is designed around rescuing people after a natural disaster.

THE GOAL:

To design and build an autonomous robot that can rescue stranded (unpainted wooden pegs) and injured victims (red wooden pegs) of a hurricane from random locations and return them to either the hospital or storm shelter depending on their color while avoiding obstacles. Extinguishing a simulated fire (back left corner) caused by downed power lines also earns bonus points. The robots will have a maximum time of 120 seconds in each of their four allotted trials. All objects, with the exception of the simulated fire, will be randomly placed for each of the four trials. The robot must begin within an 8” X 12” X 10” high size limit but may expand to any size during a trial. An Exhibit Session will precede the robot trials.

THE TRACK:

My First Impressions

Track Randomness

This is the first year that the competition has been designed with the expectation that teams will change programming between trials. Typically the track remains essentially the same during each trial and if there is any minor random change to the track, the robots typically required no alterations. The robots were more or less capable of adapting to any minor randomness in previous competitions.

The same could be done here, one could develop a robot that is completely autonomous and doesn’t require changes between trials however there is a large advantage to be gained by explicitly programming each trial.

The part of the rules that makes this year unique and entices teams to alter programming between trials is that the time bonus is based solely on the delivery of injured people to the hospital. That means that the most optimal solution, point-wise, would be to focus on delivering the red dowels to the hospital to get the most bonus points. Since they are randomly placed for each trial it would be most beneficial to program your robot to only visit the squares with the red dowels. Anything done after the injured are delivered is not affected by time aside from the hard stop limit of 120 seconds. In past years, time bonuses were awarded only once the entire objective was completed.

Opportunity for multiple robots

While I believe there is an opportunity for multiple robots I don’t feel it’s going to be a solution adopted by many teams. Multiple robots increases cost and the chance of failure by introducing more components. It can make things easier when you have multiple tasks and a single robot as the robot starts to get overcrowded and complex when designed to handle many different tasks. However, the only task that is really any different in this competition is the delivery of a racquetball to put out the fire in the back corner. Picking up of injured and uninjured people are the same so it’s not complicated to make a robot that does both of those tasks (they’re the same task) especially since you can hard-code where each dowel is before your trial. Lastly, while it might almost make sense to make two robots; one to pick up injured and one to pick up uninjured so you don’t have to do both at the same time and keep them segregated, there isn’t any benefit since the time bonus only applies to the injured. Therefore, you could run through once and deliver all of the injured people, then go back and take your time delivering uninjured people and there’s no confusion about how to store two different dowel types and keep them segregated.

Click here to get more details and the fully defined rules