US 12,493,293 B2
Multi-dimensional extended trajectory planning with changing dimensionality based on horizon steps
Maxim Likhachev, Pittsburgh, PA (US); Ellis Ratner, Oakland, CA (US); Lu Lyu, Medford, MA (US); Abhinav Srinivas Sai Kiran Kunapareddy, Sunnyvale, CA (US); Savio Joseph Pereira, Blacksburg, VA (US); Jonathan Michael Butzke, Pittsburgh, PA (US); Yunfei Xu, Milpitas, CA (US); Takashi Bando, Mountain View, CA (US); and Akihito Iwai, San Jose, CA (US)
Assigned to DENSO CORPORATION, Kariya (JP); and Waymo LLC, Mountain View, CA (US)
Filed by DENSO CORPORATION, Kariya (JP); and Waymo LLC, Mountain View, CA (US)
Filed on Mar. 28, 2022, as Appl. No. 17/706,127.
Claims priority of provisional application 63/168,565, filed on Mar. 31, 2021.
Prior Publication US 2022/0317697 A1, Oct. 6, 2022
Int. Cl. G05D 1/00 (2024.01); G06N 20/00 (2019.01)
CPC G05D 1/0223 (2013.01) [G05D 1/0088 (2013.01); G05D 1/0221 (2013.01); G06N 20/00 (2019.01)] 36 Claims
OG exemplary drawing
 
1. A multi-dimensional trajectory planning system comprising:
a memory configured to store map data, sensor data from sensors of a vehicle, and a planning application;
a planning module configured to execute the planning application to
determine a first dimensionality including a first set of dimensions for a first stage of a planning process, wherein the first set of dimensions include a plurality of variables, wherein each of the first set of dimensions are active and thus being estimated for the first stage and layer in time, and wherein the first set of dimensions includes two or more dimensions,
determine a second dimensionality including a second set of dimensions for a second stage of the planning process, wherein the second set of dimensions includes a subset of the first set of dimensions, and wherein the second stage has a lower level of dimensionality than the first stage;
based on the map data and the sensor data,
estimate first possible future states of the first set of dimensions for the first stage, and
estimate second possible future states of the second set of dimensions for the second stage based on values of the first possible future states,
wherein the first possible future states occur in the future subsequent to a current state of the vehicle at a current moment in time, and wherein the second possible future states occur in the future subsequent to the first possible future states, and
select a trajectory plan based on the second possible future states; and
an actuator module configured to drive the vehicle along a trajectory defined by the selected trajectory plan,
wherein
the term dimensionality refers to a number of active dimensions being estimated for a current stage and layer in time, and
the current stage and layer in time refers to the stage and layer in time being currently estimated, not necessarily to a current moment in time.