Robot Motion Planning Method Based on Incremental High-Dimensional Mixture Probabilistic Model

Figure 1

The Gaussian mixture model with different number of components is used to fit the probability distribution of the obstacle regions in the two-dimensional map (a), and the GMM-based collision query method is used to guide the algorithm to plan the path from the starting point to the end point. The white lines represent the feasible edges that algorithm explores, the green points are the starting point and the end point of the planning problem, and the red line represents the final path.

(a) Map

(b)

(c)

(d)

(e)

(f)