Creating the best behavior-based robot for graph traversal
To learn Behavior-Based Robotics (BBR), a simple but powerful, general strategy for robot control, we built a robot that will follow a graph lined on a flat surface until it finds its goal.
The robot consists mainly of a Raspberry Pi 2 and a Zumo chassis. The behavior is based on input form sensors such as camera, reflectance, ultrasound and proximity sensors.
This was an assignment for the subject TDT4113 CS Programming Project at Norwegian University of Science and Technology (NTNU).
The robot consistes of four different behaviors:
Go_forwardAjustStopTurn_around
The weight of each behavior depends on output from the sensors described in the paragraph above. One behavior then gets its motor recommendation sendt forward to the motors based on the arbitrators decision.
This happend each time the BBCON.run_one_timestep() is excecuted,
wich is 10 times/sec in our case.
A similar BBR model we based our work on.