US 12,242,272 B2
Path planning in mobile robots
Andrew Blake, Bristol (GB); Subramanian Ramamoorthy, Bristol (GB); Svetlin Valentinov Penkov, Bristol (GB); Majd Hawasly, Bristol (GB); Francisco Maria Girbal Eiras, Bristol (GB); Alejandro Bordallo Mico, Bristol (GB); and Alexandre Oliveira E Silva, Bristol (GB)
Assigned to Five AI Limited, Cambridge (GB)
Appl. No. 16/976,070
Filed by Five AI Limited, Bristol (GB)
PCT Filed Feb. 27, 2019, PCT No. PCT/EP2019/054911
§ 371(c)(1), (2) Date Aug. 26, 2020,
PCT Pub. No. WO2019/166518, PCT Pub. Date Sep. 6, 2019.
Claims priority of application No. 1803292 (GB), filed on Feb. 28, 2018.
Prior Publication US 2022/0107646 A1, Apr. 7, 2022
Int. Cl. G05D 1/00 (2024.01); B60W 60/00 (2020.01); G06V 20/58 (2022.01)
CPC G05D 1/0214 (2013.01) [B60W 60/0011 (2020.02); B60W 60/0027 (2020.02); G05D 1/0238 (2013.01); G05D 1/0253 (2013.01); G06V 20/58 (2022.01); B60W 2554/4042 (2020.02); B60W 2554/4043 (2020.02)] 16 Claims
OG exemplary drawing
 
1. A computer-implemented method of planning a path for a mobile robot in a presence of a set of obstacles, the method comprising:
computing for each obstacle a probabilistic obstacle position distribution;
computing at least one path-independent function as a combination of the probabilistic obstacle position distributions;
for at least one candidate path, determining an upper bound on a total probability of obstacle collision along that path, by aggregating the path-independent function based on an area defined by the candidate path and a mobile robot shape, wherein the path-independent function is independent of the candidate path;
wherein the defined area is a surface of a spacetime tube defined by the candidate path and the mobile robot shape, and the path-independent function is integrated or otherwise aggregated over that surface;
wherein the spacetime tube is determined as a set of spacetime voxels, wherein the path-independent function is determined by projecting faces of the voxels into a spacetime plane for each obstacle and integrating the probabilistic obstacle position distribution for that obstacle over each projection of the voxel face for that obstacle to compute a probability value for that voxel face, and wherein the path-independent function is aggregated over the surface of the spacetime tube by summing the probability values computed for the voxel faces; and
based on the at least one candidate path, planning a path for the mobile robot and maneuvering the mobile robot along the planned path.