A novel adaptive two-stage approach to dynamic optimal path planning of UAV in 3-D unknown environments

A novel adaptive two-stage approach to dynamic optimal path planning of UAV in 3-D unknown environments is proposed in this paper. The laser radar is exploited to incrementally build environmental maps so as to prevent the UAV from the threat of sudden obstacles. Meanwhile, by adding a vector field histogram, a temporary optimal target point is generated to make the UAV able to obtain a relatively optimal route. Furthermore, in order to prevent the UAV continuously trap into the local minimal points, a novel adaptive two-stage approach is constructed to adjust the step size and the probability of the target point. Finally, the effectiveness and feasibility of the proposed algorithm are verified in the simulation part. Compared with other algorithm, the proposed algorithm uses 31 iterations to obtain the optimal path of 61.529994 m and the planning time is just 0.32436 s.