Online Trajectory Planning Strategy for UAV in Dynamic Threats Based on Bug-Rapidly-exploring Random Tree Algorithm

In the process of UAV mission execution, trajectory planning algorithm must have the ability of real-time response to the dynamic environment. However, traditional UAV planning algorithms need to make a trade-off between the execution speed and the path quality. In this paper, a dynamic obstacle avoidance oriented online trajectory planning strategy (bug-rapidly-exploring random tree, Bug-RRT) is proposed. It combines global offline planning and local online planning through the event trigger strategy. The decision-making mode is changed with the urgency level. The level is determined by the weight of operation efficiency and path quality in evaluation, which are dynamically allocated under different environmental scenarios. In a known environment, the improved RRT and pruning strategy are used to get a global optimal trajectory. When UAV needs to avoid new obstacles, bug algorithm is used for online local planning. Simulation results show that this strategy can effectively improve the operation speed and stability.