CN-121994231-A - AGV path planning method and device and electronic equipment

CN121994231ACN 121994231 ACN121994231 ACN 121994231ACN-121994231-A

Abstract

The application provides an AGV path planning method, an AGV path planning device and electronic equipment, and relates to the technical field of automation. The method comprises the steps of obtaining an initial path by adopting a dynamic window method to simulate a track, merging adjacent path sections in the initial path to obtain at least one first path section, segmenting the first path section according to the curvature entropy value and the obstacle density of the first path section to obtain a plurality of second path sections, and carrying out interpolation processing on any one of the plurality of second path sections to obtain a target path of the AGV. By the method, the flexibility of AGV path planning can be improved, so that the AGV can adapt to different environmental conditions, and the navigation performance of the AGV is improved.

Inventors

WANG HAORAN
YU JIAMENG
WANG PENGJU
HE YUJING
YUAN XUANXUAN
LI ZONGHAN
XIE ZHIBIN
HUANG YUHONG

Assignees

广东电网有限责任公司惠州供电局

Dates

Publication Date: 20260508
Application Date: 20260108

Claims (10)

1. An AGV path planning method, comprising: Simulating a track by adopting a dynamic window method to obtain an initial path; Combining adjacent path segments in the initial path to obtain at least one first path segment; segmenting the first path segment according to the curvature entropy value and the obstacle density of the first path segment to obtain a plurality of second path segments; And carrying out interpolation processing on any one of the plurality of second path segments to obtain a target path of the AGV.
2. The method of claim 1, wherein the segmenting the first path segment based on the curvature entropy value and the obstacle density of the first path segment to obtain a plurality of second path segments comprises: Determining a first curvature of a path node in a first path segment for any one of the plurality of first path segments; Determining a curvature entropy value of the first path segment according to the first curvature; determining the density of the barriers according to the number of the barriers in the target area where the first path section is located based on environmental information; And if abrupt points exist in the curvature entropy value and/or the obstacle density exceeds an obstacle density threshold, segmenting the first path segment to obtain the plurality of second path segments.
3. The method of claim 2, wherein said determining the entropy of curvature of the first path segment from the first curvature comprises: Determining curvature sequences of all adjacent path nodes in the first path section according to the first curvature; determining the absolute value of the adjacent curvature difference value in the curvature sequence according to the curvature sequence; Determining the local entropy of the path node in the first path segment according to the absolute value of the adjacent curvature difference value; And determining the curvature entropy value of the first path segment according to the local entropy.
4. A method according to any one of claims 1 to 3, wherein said merging adjacent path segments in said initial path to obtain at least one first path segment comprises: and merging adjacent path segments in the initial path according to at least one preset merging condition to obtain at least one first path segment: The curvature change directions of the segmentation points of the adjacent path segments in the initial path are the same; the difference value of the curvature variation of the segment points is smaller than or equal to the curvature variation threshold value; The first path segment obtained after the adjacent path segments are combined meets the kinematic constraint of the AGV; the shortest distance between the first path segment and the obstacle is less than or equal to a safe distance threshold.
5. A method according to any one of claims 1 to 3 wherein interpolating any one of the plurality of second path segments to obtain a target path for the AGV comprises: determining, for any one of the plurality of second path segments, a second curvature of a path node in the second path segment; Determining an intermediate control point in the second path segment according to the path node corresponding to the maximum value in the second curvature; Performing interpolation processing on the second path segment according to the intermediate control point based on preset interpolation parameters to obtain a Bezier curve corresponding to the second path segment; and merging the Bezier curves corresponding to the second path segments respectively to obtain a target path of the AGV.
6. The method of claim 5, wherein determining an intermediate control point in the second path segment from the path node corresponding to the maximum value in the second curvature comprises: if a second-order Bezier curve is adopted for path smoothing, determining a path node corresponding to the maximum value in the second curvature as an intermediate control point in the second path segment; And if the path smoothing is carried out by adopting a third-order Bezier curve, determining the path nodes corresponding to the maximum value and the next-largest value in the second curvature as the middle control points in the second path section.
7. A method according to any one of claims 1 to 3, wherein said simulating a trajectory using a dynamic window method to obtain an initial path comprises: Optimizing weights in corresponding evaluation functions of candidate speed pairs in a dynamic window range through a differential evolution algorithm to obtain target weights, wherein the evaluation functions score the candidate speed pairs based on path approximation, safety and speed, and the dynamic window range meets the kinematic constraint of an AGV; and determining the predicted track corresponding to the target weight as the initial path.
8. The method of claim 7, wherein optimizing the weights in the corresponding evaluation functions for candidate speeds in the dynamic window range by the differential evolution algorithm to obtain the target weights comprises: according to the weight coefficient in the evaluation function, determining a weight vector and a variation vector corresponding to the weight coefficient; Based on a preset fusion probability and a component index corresponding to the weight in the evaluation function, fusing the weight vector and the variation vector to obtain a test vector; Determining a more optimal target individual in the test vector and the weight vector based on the evaluation function; And determining the weight corresponding to the target individual as the target weight.
9. An AGV path planning apparatus comprising: The simulation module is used for simulating the track by adopting a dynamic window method to obtain an initial path; The merging module is used for merging adjacent path segments in the initial path to obtain at least one first path segment; the segmentation module is used for segmenting the first path segment according to the curvature entropy value and the obstacle density of the first path segment to obtain a plurality of second path segments; And the interpolation module is used for carrying out interpolation processing on any one of the plurality of second path segments to obtain a target path of the AGV.
10. An electronic device is characterized by comprising a memory and a processor; The memory stores computer-executable instructions; The processor executing computer-executable instructions stored in the memory, causing the processor to perform the method of any one of claims 1 to 8.

Description

AGV path planning method and device and electronic equipment Technical Field The present application relates to the field of automation technologies, and in particular, to an AGV path planning method, an apparatus, and an electronic device. Background Along with the continuous promotion of intelligent substation project, automatic guided vehicle (Automated Guided Vehicle, AGV) is as an important autonomous mobile device, plays important role in tasks such as substation equipment state monitoring, meter photographing and infrared temperature measurement, can help the substation operation and maintenance personnel to grasp equipment health condition in real time and in time discover potential fault hidden danger. However, the layout of the substation facilities is complex, and factors such as narrow space between high-voltage devices, frequent changes of security fences and mobility of staff can affect the movement of the AGVs. The traditional AGV path planning method often depends on fixed environment data and a simple obstacle avoidance algorithm, so that the change of obstacles is difficult to effectively cope with in a complex and dynamic environment of a transformer substation, the path planning is not flexible enough, the obstacle avoidance reaction is not timely, and the situation such as collision or in-situ winding can possibly happen in severe cases, so that tasks cannot be executed. Disclosure of Invention The embodiment of the application provides an AGV path planning method, an AGV path planning device and electronic equipment, which are used for improving the flexibility of AGV path planning and enabling an AGV to adapt to the effect of complex environments. In a first aspect, an embodiment of the present application provides an AGV path planning method, including: Simulating a track by adopting a dynamic window method to obtain an initial path; merging adjacent path segments in the initial path to obtain at least one first path segment; segmenting the first path segment according to the curvature entropy value and the obstacle density of the first path segment to obtain a plurality of second path segments; and carrying out interpolation processing on any one of the plurality of second path segments to obtain a target path of the AGV. In one possible embodiment, the segmentation of the first path segment according to the curvature entropy value and the obstacle density of the first path segment, to obtain a plurality of second path segments, includes: determining a first curvature of a path node in a first path segment for any one of a plurality of first path segments; Determining a curvature entropy value of the first path section according to the first curvature; determining the density of the barriers according to the number of the barriers in the target area where the first path section is located based on the environmental information; If abrupt points exist in the curvature entropy value and/or the obstacle density exceeds an obstacle density threshold, segmenting the first path segment to obtain a plurality of second path segments. In one possible embodiment, determining the curvature entropy value of the first path segment from the first curvature comprises: Determining curvature sequences of all adjacent path nodes in the first path section according to the first curvature; Determining the absolute value of the adjacent curvature difference value in the curvature sequence according to the curvature sequence; Determining the local entropy of the path node in the first path section according to the absolute value of the adjacent curvature difference value; And determining the curvature entropy value of the first path segment according to the local entropy. In one possible implementation manner, merging adjacent path segments in the initial path to obtain at least one first path segment includes: combining adjacent path segments in the initial path according to at least one preset combining condition to obtain at least one first path segment: The curvature change directions of the segmentation points of the adjacent path segments in the initial path are the same; the difference value of the curvature variation of the segment points is smaller than or equal to the curvature variation threshold value; The first path segment obtained after the adjacent path segments are combined meets the kinematic constraint of the AGV; the shortest distance between the first path segment and the obstacle is less than or equal to a safe distance threshold. In one possible implementation manner, interpolation processing is performed on any one of the plurality of second path segments to obtain a target path of the AGV, including: Determining, for any one of the plurality of second path segments, a second curvature of a path node in the second path segment; determining an intermediate control point in the second path segment according to the path node corresponding to the maximum value in the second curvature;