Motion Planning
Guided Sampling-Based Motion Planning with Dynamics in Unknown Environments. This project addresses the challenges of computing dynamically feasible trajectories in unknown environments since constantly replanning as unseen obstacles are revealed during navigation both incurs significant computational expense and can introduce problematic oscillatory behavior. To improve the quality of motion planning in partial maps, this project develops a framework that augments sampling-based motion planning to leverage a high-level discrete layer and prior solutions to guide motion-tree expansion during replanning, affording both (i) faster planning and (ii) improved solution coherence. Our framework shows significant improvements in runtime and solution distance when compared with other sampling-based motion planners. (view the paper — CASE 2023)
Learning-Guided Runtime-Prediction Motion Planning This project addresses the challenge of improving the efficiency of sampling-based motion planners by incorporating machine learning to predict the expected runtime of a planner. We developed a Learning-based Motion-Planning Runtime Prediction to guide the search toward regions with low expected runtime prediction resulting in 30% shorter travel distance, up to 3x faster of planning time comparing to state-of-the-art sampling-based motion planners. (view the paper — IROS 2022)
Trajectory Optimization for AUBO Manipulator for object sorting. Utilization of Trajopt to generate trajectories for AUBO manipulator as a core component for our sorting application. (view the paper — IRC 2020)