VEHICLE TRACK PLANNING METHOD, DEVICE, COMPUTER DEVICE AND COMPUTER-READABLE STORAGE MEDIUM
A vehicle track planning method, device are provided. The method includes: dividing a road scene from an origin to a destination into a plurality of grids, wherein a grid with an obstacle and a grid without obstacle are identified with respective scene information; constructing a plurality of functions B=f (A, W) of the scene information A for identifying a grid and a planning strategy B, wherein W represents a neural network model W, and the planning strategy B comprises information of each position point in the grids through which a planning path from the origin to the destination passes; fitting the plurality of constructed planning functions B=f (A, W) to obtain the neural network model W; and obtaining a planning track from the origin to the destination according to the neural network model W.
1 . A vehicle track planning method, comprising:
dividing a road scene from an origin to a destination into a plurality of grids, wherein a grid with an obstacle and a grid without obstacle are identified with respective scene information;
constructing a plurality of functions B=f (A, W) of the scene information A for identifying a grid and a planning strategy B, wherein W represents a neural network model W, and the planning strategy B comprises information of each position point in the grids through which a planning path from the origin to the destination passes;
fitting the plurality of constructed planning functions B=f (A, W) to obtain the neural network model W; and
obtaining a planning track from the origin to the destination according to the neural network model W.
2 . The vehicle track planning method according to claim 1 , wherein the dividing a road scene from an origin to a destination into a plurality of grids comprises: numbering each of the plurality of grids.
3 . The vehicle track planning method according to claim 2 , wherein a grid with an obstacle and a grid without obstacle are identified with respective scene information by:
identifying the grid with the obstacle by scene information comprising a type and a state of the obstacle, and identifying the grid without obstacle by identical scene information.
4 . The vehicle track planning method according to claim 3 , wherein the constructing a plurality of functions B=f (A, W) of the scene information A identifying a grid and a planning strategy B comprises:
for each grid through which a planning path from the origin to the destination passes, constructing a function B=f (A, W) of the scene information A identifying the grid and the planning strategy B according to a specific condition.
5 . The vehicle track planning method according to claim 4 , wherein the specific condition comprises a shortest time, a shortest distance, an expressway priority, and/or avoidance of congestion.
6 . The vehicle track planning method according to claim 5 , wherein the information of each position point in the grids through which a planning path from the origin to the destination passes comprises:
an abscissa and an ordinate of a specific point in the grid.
7 . A vehicle track planning device, comprising:
one or more processors; and
a storage device configured for storing one or more programs, wherein
the one or more programs are executed by the one or more processors to enable the one or more processors to:
divide a road scene from an origin to a destination into a plurality of grids, wherein a grid with an obstacle and a grid without obstacle are identified with respective scene information;
construct a plurality of functions B=f (A, W) of the scene information A for identifying a grid and a planning strategy B, wherein W represents a neural network model W, and the planning strategy B comprises information of each position point in the grids through which a planning path from the origin to the destination passes;
fit the plurality of constructed planning functions B=f (A, W) to obtain the neural network model W; and
obtain a planning track from the origin to the destination according to the neural network model W.
8 . The vehicle track planning device according to claim 7 , wherein the one or more programs are executed by the one or more processors to enable the one or more processors further to: number each of the plurality of grids.
9 . The vehicle track planning device according to claim 8 , wherein a grid with an obstacle and a grid without obstacle are identified with respective scene information by:
identifying the grid with the obstacle by scene information comprising a type and a state of the obstacle, and identifying the grid without obstacle by identical scene information.
10 . The vehicle track planning device according to claim 9 , wherein the one or more programs are executed by the one or more processors to enable the one or more processors further to:
for each grid through which a planning path from the origin to the destination passes, construct a function B=f (A, W) of the scene information A identifying the grid and the planning strategy B according to a specific condition.
11 . The vehicle track planning device according to claim 10 , wherein the specific condition comprises a shortest time, a shortest distance, an expressway priority, and/or avoidance of congestion.
12 . The vehicle track planning device according to claim 11 , wherein the information of each position point in the grids through which a planning path from the origin to the destination passes comprises:
an abscissa and an ordinate of a specific point in the grid.
13 . A non-volatile computer-readable storage medium, storing computer executable instructions stored thereon, that when executed by a processor cause the processor to perform operations comprising:
dividing a road scene from an origin to a destination into a plurality of grids, wherein a grid with an obstacle and a grid without obstacle are identified with respective scene information;
constructing a plurality of functions B=f (A, W) of the scene information A for identifying a grid and a planning strategy B, wherein W represents a neural network model W, and the planning strategy B comprises information of each position point in the grids through which a planning path from the origin to the destination passes;
fitting the plurality of constructed planning functions B=f (A, W) to obtain the neural network model W; and
obtaining a planning track from the origin to the destination according to the neural network model W.
14 . The non-volatile computer-readable storage medium of claim 13 , wherein the computer executable instructions, when executed by a processor, cause the processor to perform further operations comprising: numbering each of the plurality of grids.
15 . The non-volatile computer-readable storage medium of claim 14 , wherein a grid with an obstacle and a grid without obstacle are identified with respective scene information by:
identifying the grid with the obstacle by scene information comprising a type and a state of the obstacle, and identifying the grid without obstacle by identical scene information.
16 . The non-volatile computer-readable storage medium of claim 15 , wherein the computer executable instructions, when executed by a processor, cause the processor to perform further operations comprising:
for each grid through which a planning path from the origin to the destination passes, constructing a function B=f (A, W) of the scene information A identifying the grid and the planning strategy B according to a specific condition.
17 . The non-volatile computer-readable storage medium of claim 16 , wherein the specific condition comprises a shortest time, a shortest distance, an expressway priority, and/or avoidance of congestion.
18 . The non-volatile computer-readable storage medium of claim 17 , wherein the information of each position point in the grids through which a planning path from the origin to the destination passes comprises:
an abscissa and an ordinate of a specific point in the grid.