PythonVehicleControl

Designing control systems is always an iterative process, where you as an engineer is constantly trying to improve the ways in which you can have a sandboxed environment for prototyping. While working on lateral control software for self-driving cars, I had a pain-point about not being able to easily prototype and visualize the performance of the control algorithms and compare them through multiple runs of simulations. PythonVehicleControl is a project I used to evaluate the python ecosystem for constructing a quick-and-dirty simulation environment. Several algorithms ( PID, MPC and DDPG RL) are implemented and GUI/visualization is done using PyQT and PyQTGraph.