CPC G06V 10/764 (2022.01) [G01C 21/3407 (2013.01); G01C 21/3453 (2013.01); G06N 3/044 (2023.01); G06N 3/0455 (2023.01); G06V 10/82 (2022.01); G06V 20/56 (2022.01)] | 18 Claims |
1. A trajectory prediction method, comprising:
acquiring current trajectory data and current map data, wherein the current trajectory data comprises a plurality of trajectory points of a moving subject performing movement in current environment and one or more other moving objects in the current environment in a past predetermined time period, each of the plurality of trajectory points comprises a spatial position at a corresponding time point, and the current map data comprises spatial positions of a plurality of map points of a road element in the current environment in the predetermined time period;
expressing the current trajectory data and the current map data as a current trajectory point set and a current map point set in a high-dimensional space;
extracting a global scene feature according to the current trajectory point set and the current map point set;
predicting a plurality of prediction trajectory point sets of the moving subject and a probability corresponding to each of the plurality of prediction trajectory point sets according to the global scene feature; and
controlling a vehicle to travel along a path planned according to the plurality of prediction trajectory point sets and the probability corresponding to each of the plurality of prediction trajectory point sets.
|