This repository contains all (ongoing) research developmental work on Adaptive Deferred-Decision Trajectory Optimization (Adaptive-DDTO), a robust path-planning algorithm which maintains multiple targets in the reachable set of the ego vehicle and adaptively recomputes once the target count falls below a preset threshold. An implementation of Adaptive-DDTO for the quadcopter landing problem (as described in the HALO research paper, see below) is provided in the 📂src folder. Please see the following papers for more context:
- HALO: Hazard-Aware Landing Optimization for Autonomous Systems
- Deferring Decision in Multi-target Trajectory Optimization (along with this presentation)
This repository contains entirely Julia code, with associated Manifest/Project files at the top level (for more information on how Julia works, see the Julia tutorials page). All code is contained within the 📂src folder, which notably contains a 📜demo.ipynb jupyter notebook file that demonstrates the capabilities of Adaptive-DDTO on a sandbox problem, along with all associated solving, plotting and utility files. Currently, you will need to have all packages mentioned in 📜src/setup.jl for both internal use in this package, and for use in the HALO framework as a submodule.
If you use Adaptive-DDTO in your research, kindly cite the following associated publication.
@inproceedings{hayner2023halo,
title={HALO: Hazard-Aware Landing Optimization for Autonomous Systems},
author={Hayner, Christopher R and Buckner, Samuel C and Broyles, Daniel and Madewell, Evelyn and Leung, Karen and A{\c{c}}ikme{\c{s}}e, Beh{\c{c}}et},
booktitle={2023 IEEE International Conference on Robotics and Automation (ICRA)},
pages={3261--3267},
year={2023},
organization={IEEE}
}