This repository includes the MATLAB code used to simulate quadrotors and perform Reachability-based Trajectory Design (RTD) for our ASME DSCC 2019 paper [1], which won the best student paper prize.
RTD is a method for safe online trajectory planning for mobile robots. Check out the RTD repo (for car-like robots) and the RTD tutorial.
You'll need the following repositories on your MATLAB path:
Run the script run_quadrotor_simulation_static_obstacles.m in the simulation folder.
Planning with RTD happens in the quadrotor_RTD_zono_planner.m class, which gets called by the simulator (from the repo of the same name).
If you want to walk through the reachable set computation, start at step_1_frs_computation; note that computing the tracking error tables in step 2 can take several hours, so we have provided some precomputed tracking error tables for you.
Shreyas Kousik, Patrick Holmes, and Zehui Lu
[1] Kousik, S., Holmes, P. and Vasudevan, R., 2019, October. Safe, aggressive quadrotor flight via reachability-based trajectory design. In Dynamic Systems and Control Conference (Vol. 59162, p. V003T19A010). American Society of Mechanical Engineers. link
[2] Kousik, S., Holmes, P. and Vasudevan, R., 2019. Technical Report: Safe, Aggressive Quadrotor Flight via Reachability-based Trajectory Design. arXiv preprint arXiv:1904.05728. link