This repository contains code for trajectory optimization for an omnidirectional tire-wheeled robot. Complete system dynamics are available, with a wrapper to model any constraints into the optimization scheme. All research was done as part of my master thesis in robotic engineering, the report of which can be found here.
The complete kinematic and dynamic models for the Otbot can be found in the folder Otbot-models. The Results and model folder must be added to path before running.
A wrapper of the OptimTraj library for the Otbot can be found in the OtbotTrajOpt/Otbot folder. It includes everything necessary to create any desired scenario and comptue optimal trajectories using direct collocation as implemented in OptimTraj.
To compute a trajectory, simply run the MAIN_loop.m with your desired parameters. This function is designed to run several optimizations in a row, for each set of parameters provided. A list of possible plots is also available.
To model obstacles, use the files in the Obstacles folder. The number of obstacles must be provided accurately for the optimization to run.
To make videos of simulation, use the RESULTS_makeplots.m function. Simply provide the name of your solution and put 'yes' for the make_video variable. A list of possibe plots is also available.
Current folder must be Otbot-TrajOpt when running to save files in correct emplacement. OptimTraj and OTbot folder must both be added to matlab path. All results will be saved in the Otbot/Results folder.