Research

We propose a risk-adaptive approach that combines the Conditional Value-at-Risk Barrier Function Constraint, which automatically adjusts risk to balance safety and feasibility, with a dynamic zone-based barrier function that evaluates collision likelihood from relative states, enabling adaptive safety margin expansion and proactive obstacle avoidance.

We propose a spatial decomposition based coverage control that mitigates the issue of local minima trapping in general unstructured obstacle environments supported by convergence guarantees.

We propose a novel pursuit control policy that utilizes the Voronoi partition method to collectively encircle the evader while ensuring collision avoidance, ultimately leading to its capture.

We propose a fully autonomous infrastructure inspection and information management system based on multi-agent coverage control method