In this project I designed a RRT path planner for a differential drive robot and implemented an optimization-based path planner, similar to MPC. I worked alongside my partner to implement an additional path planner that steers with sinusoids based off this paper [1].
Unlike previous path planning projects that assume omni-directional control (such as the RoboCup robots), all the planners designed or implemented in this project take system dynamics into account. The model and dynamics are based off of Ackermann steering shown below.
The results are described in more detail in the report. The optimization-based planner understandably produced the most optimal paths, however, took significant time to do so. The RRT planner can quickly generate paths, however, the paths could benefit from additional constraints to smooth them out.
Learned how to incorporate dynamics into path planning.