CN-122018496-A - Road condition perception fused AGV bidirectional traffic obstacle avoidance track generation method and system

CN122018496ACN 122018496 ACN122018496 ACN 122018496ACN-122018496-A

Abstract

The invention provides a road condition perception fused AGV bidirectional traffic obstacle avoidance track generation method and system, and relates to the technical field of track planning, wherein the method comprises the following steps: the ground condition is scanned in real time through the vehicle-mounted laser radar, the driving interference is identified, the safety speed is fitted, the dynamic safety distance is calculated based on the real-time safety speed of two AGVs, if the collision distance is smaller than the safety distance, the obstacle avoidance data acquisition is triggered, the safety area is shared and avoided by calculating the asymmetric safety turning capacity, the simulated driving route is generated, the role distribution is carried out according to the energy consumption prediction, the driving track of the driving vehicle is determined and controlled, meanwhile, the speed of the non-driving vehicle is reduced, the safety avoidance of the two AGVs is ensured, and the driving is carried out until the driving is away from the shared driving area. The invention solves the technical problems that the prior art only depends on static path planning to realize bidirectional passing obstacle avoidance, which causes the AGV to be difficult to flexibly cope in a complex environment, and further causes the reduction of running efficiency and safety.

Inventors

LIANG YE
LIANG MIN
YAO JIANDONG
KANG YUXIA
HUANG CHAOXIONG

Assignees

广西玉柴机器股份有限公司

Dates

Publication Date: 20260512
Application Date: 20250929

Claims (9)

1. The method for generating the bidirectional passing obstacle avoidance track of the AGV integrating road condition perception is characterized by comprising the following steps of: Carrying out driving interference identification according to a first real-time ground condition obtained by the first AGV through vehicle-mounted laser radar scanning, and fitting and outputting a first real-time safe speed; Carrying out driving interference identification according to a second real-time ground condition obtained by the second AGV through vehicle-mounted laser radar scanning, and fitting and outputting a second real-time safety speed; Solving a dynamic safety distance according to the first real-time safety speed and the second real-time safety speed; If the real-time conflict distance is smaller than the dynamic safety distance, triggering the first AGV and the second AGV to acquire first obstacle avoidance associated data and second obstacle avoidance associated data respectively; Calculating asymmetric safe turning capacity based on the first obstacle avoidance associated data and the second obstacle avoidance associated data, and positioning a shared avoidance safety zone; Taking the shared avoidance safety zone as a yield boundary constraint, and fitting and outputting a first simulated yield route and a second simulated yield route; Predicting the first yield energy consumption and the second yield energy consumption of the first simulation yield route and the second simulation yield route respectively by taking the yield direction as constraint; Performing role allocation of the first AGV and the second AGV by comparing the first yielding energy consumption and the second yielding energy consumption, and positioning a real-time yielding route of the real-time yielding vehicle and an original driving track of the non-yielding vehicle; And controlling the non-yielding vehicles to run at a reduced speed on the original driving track in the running process of controlling the real-time yielding vehicles according to the real-time yielding route until the real-time yielding vehicles and the non-yielding vehicles leave the shared avoidance safety zone.
2. The method for generating the bidirectional traffic obstacle avoidance trajectory of the AGV fused with road condition awareness as set forth in claim 1, further comprising: the method comprises the steps of calling a first preset track of a first AGV and a second preset track of a second AGV from a workshop logistics table; Acquiring a first real-time position of the first AGV and a second real-time position of the second AGV through an interactive vehicle UWB positioning system; and calculating the real-time conflict distance according to the first real-time position and the second real-time position by taking a track intersection section of the first preset track and the second preset track as a constraint.
3. The method for generating the bidirectional traffic obstacle avoidance trajectory of the AGV fused with road condition awareness according to claim 2, wherein the performing the driving interference recognition according to the first real-time ground condition obtained by the first AGV through the vehicle-mounted laser radar scanning, fitting and outputting the first real-time safe speed comprises: the vehicle-mounted laser radar of the first AGV scans the ground of a channel at the frequency of 10Hz to acquire first reflection intensity point cloud data; Based on the reflectivity distribution characteristics, carrying out driving interference feature identification on the first reflection intensity point cloud data by adopting a double-threshold segmentation method to obtain a first oil pollution distribution area and a first metal debris distribution area; And after the first oil stain distribution area and the first metal debris distribution area are spatially overlapped, carrying out adhesion safety analysis, and fitting and outputting the first real-time safety speed.
4. The method for generating the bidirectional traffic obstacle avoidance trajectory of the AGV fused with road condition awareness according to claim 3, wherein after spatially overlapping the first oil pollution distribution area and the first metal debris distribution area, performing adhesion safety analysis, and fitting and outputting the first real-time safety speed comprises: The first real-time transport load returned by the load sensor of the first AGV is accessed; Receiving a first real-time gradient angle and a first real-time gradient direction output by an IMU of the first AGV; Performing safe adhesiveness attenuation analysis based on the first oil stain distribution area and the first metal debris distribution area to obtain a first oil stain attenuation coefficient and a first debris disturbance coefficient; After the first oil stain distribution area and the first metal debris distribution area are spatially overlapped, a dynamic attachment model is built by combining the first real-time gradient angle; Based on the track segment of the first preset track at the first real-time position, carrying out adhesion safety analysis on the dynamic adhesion model, and outputting an initial safety speed; And compensating the initial safety speed by adopting the first real-time gradient direction, and outputting the first real-time safety speed.
5. The traffic perception fusion AGV bidirectional traffic obstacle avoidance trajectory generation method according to claim 3, wherein the performing driving interference feature recognition on the first reflection intensity point cloud data by adopting a dual-threshold segmentation method based on reflectivity distribution characteristics to obtain a first oil pollution distribution area and a first metal debris distribution area comprises: Removing invalid points with the reflection intensity of less than 5%, and filtering background noise of the first reflection intensity point cloud data; Traversing the first reflection intensity point cloud data to extract discrete point cloud clusters with the reflection intensity of more than 85%; traversing the discrete point cloud clusters by adopting a preset point cloud space density threshold to remove isolated high-light points, and outputting the first metal debris distribution area; traversing the first reflection intensity point cloud data to extract continuous point cloud clusters with reflection intensity of 5% -30%; after the micro-area elimination is carried out by traversing the continuous point cloud clusters through a preset point cloud projection area threshold, the adjacent greasy dirt spots are fused by carrying out morphological closing operation on the continuous point cloud clusters, and the first greasy dirt distribution area is generated.
6. The method for generating the bidirectional traffic obstacle avoidance trajectory of the AGV fused with road condition awareness according to claim 4, wherein solving the dynamic safety distance according to the first real-time safety speed and the second real-time safety speed comprises: Taking the first real-time gradient direction as a deceleration correction constraint, pre-calculating a first deceleration capacity according to the first real-time transport load and a first real-time safe speed, and analogizing to solve a second deceleration capacity of the second AGV; outputting a dominant safety speed by comparing the first real-time safety speed with the second real-time safety speed; outputting a dominant deceleration by comparing the first deceleration capacity and the second deceleration capacity; calculating a safety reaction distance based on a preset fixed response time and the dominant safety speed; calculating a theoretical braking distance based on the dominant safe speed and dominant deceleration; matching a first dynamic environment allowance according to the first oil stain distribution area and the first metal debris distribution area; matching a second dynamic environment allowance according to the second oil stain distribution area and the second metal debris distribution area; and on the basis of adding the safety reaction distance and the theoretical braking distance, taking the maximum value of the first dynamic environment allowance and the second dynamic environment allowance to carry out safety distance compensation, and outputting the dynamic safety distance.
7. The traffic awareness fused bidirectional traffic obstacle avoidance trajectory generation method of an AGV of claim 4 wherein calculating asymmetric safe turning capacity based on the first obstacle avoidance associated data and the second obstacle avoidance associated data locates a shared avoidance safety zone comprising: when the first AGV is an ascending AGV and the second AGV is a descending AGV: Carrying out front wheel load attenuation compensation according to the first real-time transportation load, the first real-time gradient angle and the first real-time gradient direction, and outputting a first minimum turning radius; carrying out rear wheel adhesive force attenuation compensation according to the second real-time transportation load, the second real-time gradient angle and the second real-time gradient direction, and outputting a second minimum turning radius; based on an uphill expansion rule, mapping the first minimum turning radius in the first real-time position space, and generating a first dynamic envelope; based on a downhill expansion rule, mapping the second minimum turning radius in the second real-time position space to generate a second dynamic envelope; carrying out union solution on the first dynamic envelope body and the second dynamic envelope body to obtain a conflict core area; And after the physical boundary of the transportation channel is obtained interactively, performing geometric Boolean operation on the physical boundary of the transportation channel and the core conflict area through convex polygon Boolean subtraction operation, and outputting the sharing avoidance safety area.
8. The method for generating the bidirectional traffic obstacle avoidance trajectory of the AGV fused with road condition awareness according to claim 7, wherein the step of taking the shared avoidance safety zone as a yield boundary constraint to fit and output a first simulated yield route and a second simulated yield route comprises the steps of: The first preset track and the second preset track are spatially segmented by adopting the shared avoidance safety zone, so that a first segmented track and a second segmented track are obtained; Under the scene that the first segmentation track is fixed, the shared avoidance safety space is taken as a constraint, shortest obstacle avoidance track fitting is carried out, and a shortest obstacle avoidance cubic spline curve meeting the second minimum turning radius is generated and is used as the output of the second simulated yielding route; And under the scene that the second dividing track is fixed, the shared avoidance safety space is taken as a constraint, shortest obstacle avoidance track fitting is carried out, and a shortest obstacle avoidance Bezier curve meeting the first minimum turning radius is generated and is used as the output of the first simulated yielding route.
9. The road condition sensing fusion AGV bidirectional traffic obstacle avoidance trajectory generation system, which is characterized by being used for implementing the road condition sensing fusion AGV bidirectional traffic obstacle avoidance trajectory generation method according to any one of claims 1-8, wherein the system comprises: the first traveling interference recognition unit is used for carrying out traveling interference recognition according to a first real-time ground condition obtained by the first AGV through the vehicle-mounted laser radar scanning, and fitting and outputting a first real-time safe speed; the second driving interference recognition unit is used for carrying out driving interference recognition according to a second real-time ground condition obtained by the second AGV through the vehicle-mounted laser radar scanning, and fitting and outputting a second real-time safe speed; the dynamic safety distance solving unit is used for solving the dynamic safety distance according to the first real-time safety speed and the second real-time safety speed; the obstacle avoidance associated data acquisition unit is used for triggering the first AGV and the second AGV to acquire first obstacle avoidance associated data and second obstacle avoidance associated data respectively if the real-time collision distance is smaller than the dynamic safety distance; The avoidance safety zone positioning unit is used for calculating asymmetric safety turning capacity based on the first obstacle avoidance associated data and the second obstacle avoidance associated data and positioning and sharing the avoidance safety zone; the simulation yielding route output unit is used for taking the shared avoidance safety zone as a yielding boundary constraint and fitting and outputting a first simulation yielding route and a second simulation yielding route; the yield energy consumption prediction unit is used for predicting the first yield energy consumption and the second yield energy consumption of the first simulated yield route and the second simulated yield route respectively by taking the yield direction as constraint; The role allocation unit is used for performing role allocation of the first AGV and the second AGV by comparing the first yielding energy consumption and the second yielding energy consumption, and positioning a real-time yielding route of the real-time yielding vehicle and an original driving track of the non-yielding vehicle; And the vehicle control unit is used for controlling the non-yielding vehicles to run at a reduced speed on the original driving track in the running process of controlling the real-time yielding vehicles according to the real-time yielding route until the real-time yielding vehicles and the non-yielding vehicles leave the shared avoidance safety zone.

Description

Road condition perception fused AGV bidirectional traffic obstacle avoidance track generation method and system Technical Field The invention relates to the technical field of track planning, in particular to a method and a system for generating an AGV bidirectional passing obstacle avoidance track integrating road condition perception. Background AGVs (automatic guided vehicles) play a vital role in the scenes of warehouse, assembly lines and the like, and in the workshop environments of assembly plants and the like, the bidirectional passing of the AGVs is particularly important, because the environments often involve narrow passages and complex traffic conditions, and the AGVs need to be able to dynamically adjust the driving paths thereof according to real-time road conditions and surrounding environments so as to avoid collisions and ensure safe and efficient operation. However, in the prior art, obstacle avoidance is realized only by depending on a static path planning and a preset turning radius, so that when an AGV faces a complex ground condition, the running strategy of the AGV may not be adjusted in time, and the situation causes that the path planning of the AGV is not flexible enough, a new obstacle is not effectively avoided or environmental change is not dealt with, the risk of collision is increased, and further, the running efficiency and the safety are reduced. Disclosure of Invention Aiming at the defects or improvement demands of the prior art, the application provides a method and a system for generating a bidirectional passing obstacle avoidance track of an AGV (automatic guided vehicle) by fusing road condition perception, which are used for solving the technical problems that the AGV is difficult to flexibly cope in a complex environment and the running efficiency and the safety are reduced because the prior art only depends on static path planning to realize the bidirectional passing obstacle avoidance. In order to achieve the above purpose, the application provides a method and a system for generating an AGV bidirectional passing obstacle avoidance track by integrating road condition perception. The application provides a method for generating an AGV bidirectional traffic obstacle avoidance track integrating road condition perception, which comprises the following steps: The method comprises the steps of carrying out driving interference identification according to a first real-time ground condition obtained by scanning a vehicle-mounted laser radar by a first AGV, fitting and outputting a first real-time safety speed, carrying out driving interference identification according to a second real-time ground condition obtained by scanning a vehicle-mounted laser radar by a second AGV, fitting and outputting a second real-time safety speed, solving a dynamic safety distance according to the first real-time safety speed and the second real-time safety speed, triggering the first AGV and the second AGV to respectively acquire first obstacle avoidance associated data and second obstacle avoidance associated data if the real-time collision distance is smaller than the dynamic safety distance, calculating asymmetric safety turning capacity based on the first obstacle avoidance associated data and the second obstacle avoidance associated data, positioning a shared avoidance safety zone, taking the shared avoidance safety zone as a constraint of a yielding boundary, fitting and outputting a first simulated yielding route and a second simulated yielding route, respectively predicting first yielding energy consumption and second yielding energy consumption of the first simulated yielding route and the second simulated yielding route by taking the yielding direction as constraint, respectively, carrying out control on the first yielding energy consumption and the second yielding route by comparing the first yielding energy consumption and the second yielding energy consumption with the second yielding route, and the first yielding energy consumption and the second yielding vehicle in real-time, and the first yielding vehicle driving safety zone and the non-real-time driving safety zone and the vehicle driving safety zone are controlled in real time. In one embodiment, the following process is also performed: The method comprises the steps of obtaining a first preset track of a first AGV and a second preset track of a second AGV from a workshop logistics table, obtaining the first real-time position of the first AGV and the second real-time position of the second AGV through an interactive vehicle UWB positioning system, and calculating the real-time conflict distance according to the first real-time position and the second real-time position by taking a track intersection section of the first preset track and the second preset track as constraint. In one embodiment, the driving interference recognition is performed according to a first real-time ground condition obtained by the first AGV through