Real-time path generation method and device and automobile

Abstract

The invention provides a real-time path generation method and device and an automobile. The method comprises the following steps: acquiring a vehicle starting point, a vehicle target point and a grid map; the vehicle starting point, the vehicle target point and the grid map are optimized, a plurality of target road points are formed, and the target road points are directed sequence points of the grid map; according to the current position of the vehicle, a preset number of target road points closest to the current position of the vehicle are selected as optimization control points corresponding to the current position of the vehicle in the advancing direction of the vehicle; according to the current position of the vehicle, the preset vehicle running step length, the optimization control point and the nearest target road point of the optimization control point in the vehicle running direction, a multi-order Bezier curve corresponding to the optimization control point is used for calculating a target track point passing through the next moment of each preset time period; and controlling the vehicle to run to the target track point according to the target track point. According to the invention, the problems that vehicle smoothness is affected by discontinuous curvature and track characteristics are lost due to excessive smoothness in the prior art are solved.

Description

technical field

[0001] The invention relates to the technical field of automobile automatic driving, in particular to a real-time path generation method, device and automobile. Background technique

[0002] There are many methods of path planning, which are widely used in the fields of self-driving vehicles, unmanned aerial vehicles, and mobile robots. Many methods of autonomous driving path planning technology are migrated from the field of mobile robots, and combined with road network, traffic rules, vehicle self-constraints, etc. to improve and optimize. At present, most self-driving vehicles use lidar solutions to generate grid maps through perception. The results of method optimization are essentially discrete waypoints. The trajectory obtained by directly connecting the waypoints is a broken line graph, the curvature is discontinuous, and there is a large lateral angular acceleration at the waypoint, which needs to be smoothed. The algorithm of the curve interpolatio...

Examples

Embodiment Construction

[0041] In this patent, the specific implementation manner will be further described below in conjunction with the accompanying drawings and embodiments.

[0042] Such as figure 1As shown, the embodiment of the present invention provides a real-time path generation method, the method comprising:

[0043] Step S11, acquiring the starting point of the vehicle, the target point of the vehicle and the grid map.

[0044] It should be noted that in the route setting of unmanned driving, the target point of the vehicle is obtained through user input, the starting point of the vehicle is the starting point of the vehicle, and the grid map is mostly obtained through lidar perception, of course, it can also be obtained through cameras or other perception devices .

[0045] Step S12 , optimizing the starting point of the vehicle, the target point of the vehicle and the grid map to form a plurality of target waypoints, and the target waypoints are directed sequence points of the grid map...

technical field

[0001] The invention relates to the technical field of automobile automatic driving, in particular to a real-time path generation method, device and automobile. Background technique

[0002] There are many methods of path planning, which are widely used in the fields of self-driving vehicles, unmanned aerial vehicles, and mobile robots. Many methods of autonomous driving path planning technology are migrated from the field of mobile robots, and combined with road network, traffic rules, vehicle self-constraints, etc. to improve and optimize. At present, most self-driving vehicles use lidar solutions to generate grid maps through perception. The results of method optimization are essentially discrete waypoints. The trajectory obtained by directly connecting the waypoints is a broken line graph, the curvature is discontinuous, and there is a large lateral angular acceleration at the waypoint, which needs to be smoothed. The algorithm of the curve interpolatio...

Examples

Embodiment Construction

[0041] In this patent, the specific implementation manner will be further described below in conjunction with the accompanying drawings and embodiments.

[0042] Such as figure 1As shown, the embodiment of the present invention provides a real-time path generation method, the method comprising:

[0043] Step S11, acquiring the starting point of the vehicle, the target point of the vehicle and the grid map.

[0044] It should be noted that in the route setting of unmanned driving, the target point of the vehicle is obtained through user input, the starting point of the vehicle is the starting point of the vehicle, and the grid map is mostly obtained through lidar perception, of course, it can also be obtained through cameras or other perception devices .

[0045] Step S12 , optimizing the starting point of the vehicle, the target point of the vehicle and the grid map to form a plurality of target waypoints, and the target waypoints are directed sequence points of the grid map...

Images