Get ready for the 2018 competition!
This weekend, we — Elnaz and Emily — were judges at Robocup Junior Queensland for the Rescue Challenges. We oversaw two divisions: the Line Follow Rescue, and the Maze Rescue. Last year we gave you tips and tricks for the Line Follow Rescue Challenge; this year, the Maze Rescue Challenge was a new addition to Robocup Junior. From having watched many teams compete at many Maze Challenges, here are our tips and tricks for winning in 2018.
There has been an accident at a manufacturing plant. The area is too dangerous to send in a human rescue team, so we need to send in an autonomous robot to navigate the terrain and obstacles, locate the victims and drop rescue packages.
Getting the most points
To get the total number of available points, your robot will need to:
Go to all three checkpoints (grey tiles) (30 points)
Identify all five victims (coloured red, heated to 20-40 degrees, located at 7cm above the floor)
If a victim is located on a linear wall* = 10 points
If a victim located on a floating wall* = 25 points
Drop a rescue package at each victim (10 points per rescue package)
Exit the maze and stay on the exit tile for 10 seconds (bonus points)
*Walls may or may not lead to the entrance/exit. Walls that lead to the entrance/exit are called linear walls. The walls that do NOT lead to the entrance/exit are called floating walls.
Beware of black tiles! Robots can step on black tiles, but they are required to retreat backwards — all the way back onto a white or grey tile — and continue on their path on the white or grey tiles. If the robot does not do this, students will be required to pick up their robot and place it down at the last checkpoint tile and start over.
Your easiest points can be obtained by going to all three checkpoints, as this requires your robot to navigate the orthogonal maze.
Victims can be identified either by their colour (red), their temperature (20-40 degrees), or both. The teams that managed to set themselves apart from the competition were those whose robots were able to identify the victims. The more effective robots located victims using their heat patterns rather than identifying them by the colour red.
Once a robot has identified a victim, the robot is required to temporarily change its lights to red to indicate that it has identified a victim.
Dropping rescue packages
The ability to drop rescue packages was only achieved by a small number of teams. This allowed the top teams to separate themselves from the rest of the competition.
Exit bonus points
Exit bonus points only apply if victims have been identified. The exit bonus points are equal to the number of points achieved from identifying victims.
Selecting your sensors
LEGO’s EV3 robots only have four sensor ports, limiting the number of sensors that you can select. Decide which sensors your robot might need. Some ideas for effective sensors might be: a Thermo sensor, a Gyro Sensor, a Colour Sensor Up, a Colour Sensor Down, or an Ultrasonic Sensor Up.
For every tile that the robot moves forward, it should turn around in all four directions to search for victims. Before programming your robot, you need to figure out exactly how many wheel degrees are needed to travel 30cm (i.e. the size of each tile) forwards, and exactly how many wheel degrees are needed to turn at a right angle (90 degree).
After testing, create a program that effectively sticks to the LEFT WALL of the maze. This technique has at least one obvious flaw! If the turns are even slightly off 90 degrees, or the forward distance is even slightly over or under 30cm, then with every tile this small error is magnified. You need to find a way to correct the robot position and orientation every now and again. Note, if there are floating walls in the maze, this method will miss the internal corridors of the maze.
Correcting the robot position and orientation
To solve the problem we described above, you can use wheel encoders to make sure the robot moves with accuracy, and use corrections to help restore the robot to the correct position whenever possible.
Most of the robots in this competition used a technique called wall-tracking. Imagine it like this: you are in the maze. Place your left or right hand on the wall. Continue traversing through the maze, always keeping your left hand on the wall.
Unfortunately, this maze searching technique won’t work on some of the Rescue Maze layouts, since they are allowed to include floating walls. In mazes that including floating walls, following the wall will not allow the robot to visit some tiles in the centre of the maze. Another solution is needed!
Mapping the field
(Knowing where you are!)
This is a significant opportunity to make your life that much easier for the Rescue Maze. If your robot can map out rooms, and then it can use this map to help it navigate tiles that have already been visited, or help it return to the exit tile to get the exit bonus.
Size of the robot
The size of your robot can be an issue. Try to design a robot which can easily traverse the maze.
Delivering a rescue package
In the competition, different groups used two different mechanism to activate the rescue package dispenser when a robot detected a victim:
1. Using a Conveyer Belt
Some robots used a conveyor belt to move each rescue package forward, until it falls off the front. Using the LEGO EV3 tracks with 3M friction pegs, it is possible to squeeze rescue packages into into each section; these release when the belt flows around and under the track system. Here’s a picture to show you what we mean:
2. Using an Actuator
Another method which was used was the chute, which essentially ‘pushed’ rescue packages out of the chute with an actuator:
Let us know what you think. Do you have any tips and tricks for the new Maze Rescue Challenge at Robocup Junior? What did you learn from you Robocup Junior experience.
Contact us on email@example.com if you are looking for robotics or STEM coaches for your school competition teams.
By Emily de la Peña and Elnaz Karimpour
Elnaz Karimpour is a mechatronics engineer with a Masters of Electrical Engineering (Mechatronics and Automation Control). She is a Coding Mentor with Coding Kids and coaches student teams for robotics and tech competitions.
Emily de la Peña is the Founder of Coding Kids and Advance Queensland’s Community Digital Champion. She is a civil engineer by trade and entrepreneur at heart.