Brief Introduction:

This work aims to plan risk-aware trajectories for MAVs in dynamic environments efficiently. The risk is to evaluate the trajectory’s safety, considering the time dimention, and is defined for each trajectory by leveraging the particles in our DSP map. The MAV with our planner can fly up to 2.5 m/s in the environment with both static obstalces and dynamc obstacles.

To the best of the authors’ knowledge, this is the firstquadrotor obstacle avoidance system that can avoid arbitrary-shaped dynamic and static obstacles simultaneously. The contributions of this paper include:

• An efficient risk-aware sampling approach composedof two-phase motion primitives in the state space.
• The approach to merge the local trajectory given byour planner to an arbitrary global trajectory.
• A complete obstacle avoidance system that can beapplied to light-weight quadrotors in dynamic environ-ments.

This paper is accepted by IEEE Transactions on Industrial Electronics. https://ieeexplore.ieee.org/document/10034468.