Based on the paper Optimal Path Planning of a Mobile Robot using Quadrant Based Random Particle Optimization Method
Was originally ported to Arduino but I'm currently testing it on MATLAB, so the Arduino port is outdated.
The code shows cartesian coordinates for the robot to move while avoiding a single obstacle anywhere in a static environment. This can of course easily be scaled to navigate in a dynamic environment with multiple obstacles. This code DOES NOT operate a robot right away, it simply takes environment properties as inputs and gives coordinates as output.
- RPO cpp folder contains the C++ implementation
- RPO MATLAB contains a matlab script and functions, currently being worked upon
- Trajectory, obstacle, and goal plotting
- The algorithm itself, but in very limited situations
- The algorithm itself, in most situations
- Wildly different response when the number of artificial points or step-size is changed
- Recheck the math behind gaussian potential calculations, since the paper uses a slightly modified version of a 2D gaussian
- Plot APs on each iteration along with their potentials, to recheck the AP selection criteria
- Implement changes in Arduino and cpp, once the algo works