CN-121977598-A - Engineering truck group dynamic path planning method for intelligent construction site

CN121977598ACN 121977598 ACN121977598 ACN 121977598ACN-121977598-A

Abstract

The invention discloses a dynamic path planning method for an engineering truck group in an intelligent construction site, which is used for realizing efficient path planning and dynamic obstacle avoidance of engineering trucks in a complex environment by fusing an A-scale algorithm and a dynamic artificial potential field method. Firstly, generating a local optimal path for each engineering vehicle based on two-dimensional grid environment modeling by adopting a multi-node expansion A-type algorithm, then detecting and comprehensively solving path conflicts among different engineering vehicles through constraint trees, eliminating redundant nodes through inflection point optimization strategies, shortening the path length, improving smoothness and further generating a global optimal path, detecting moving obstacles in real time by combining a multi-sensor fusion technology in a dynamic obstacle avoidance stage, judging collision risks through calculating object area change rates and distance thresholds, introducing a relative speed repulsive potential field to improve a traditional artificial potential field model when the collision risks exist, dynamically correcting the path direction, controlling the engineering vehicles to decelerate and avoid, solving the problem of predicting the motion direction of the dynamic obstacles, and guiding the engineering vehicles to return to the original path with the minimum deviation cost through a track recovery algorithm after obstacle avoidance is completed. The method effectively solves the problems of path redundancy, dynamic obstacle avoidance response lag and artificial potential field local oscillation of the traditional A-based algorithm, has the characteristics of high path global optimality, fast dynamic obstacle avoidance response and low hardware cost, and is suitable for engineering vehicle operation scenes in intelligent construction sites.

Inventors

ZHANG XUEJIE
WU SHUBIN
WANG LONGBAO
YING HAIHUA
LIU FAN
HUANG XUANXUAN
ZHOU ZHIJUN

Assignees

河海大学

Dates

Publication Date: 20260505
Application Date: 20260121

Claims (8)

1. The engineering truck group dynamic path planning method for the intelligent construction site is characterized by comprising the following steps of: s1, modeling a site area; s2, planning a bicycle path for each engineering truck; s3, detecting and comprehensively solving path conflicts among different engineering vehicles; S4, extracting a global path inflection point as a virtual target point and judging whether the global path inflection point is a critical path, wherein the extracted critical path is the optimal path of the current engineering truck; s5, judging whether the engineering vehicle is likely to collide with the moving object or not according to the acquired image and the ranging result; s6, if collision is not possible, continuing running, if collision is possible, decelerating the engineering vehicle and adding a speed repulsive potential field to plan a dynamic path again; and S7, returning the engineering truck to the original path through a track recovery algorithm after obstacle avoidance is completed.
2. The method for dynamic path planning of engineering truck group for intelligent construction site according to claim 1, wherein the step S1 is specifically: firstly, modeling a space environment, dividing the searched space environment into grids with equal size, wherein each grid is called a unit, the position of the grid is determined by rows and columns of the grid, the search area is stored in a two-dimensional array, each item of the array represents one grid, the state space of the array is walkable or non-walkable, the walkable grid has no obstacle, the engineering truck can pass through the grid at the moment, and the non-walkable grid has an obstacle, and the engineering truck can not pass through the grid at the moment.
3. The method for managing an engineering vehicle for an intelligent construction site according to claim 1, wherein the step S2 is specifically: a1, taking a starting point as A1 st node, calculating a cost function of an expansion node, continuously taking a node with the minimum cost value as a child node in the expansion process, taking the child node as a current father node to expand and search the child node in the next process, and sequentially searching out the child node with the minimum cost function to form an optimized track; the formula of the evaluation function A2:F (n) is as follows: F(n)=G(n)+H(n) f (n) represents the total cost from the starting point to the target point through the node n, G (n) represents the actual cost from the starting point to the current node n, and the calculation formula is as follows: H (n) is a heuristic estimated cost from the current node n to the endpoint, and its calculation formula is: Wherein (x n ,y n ) is the current node (x s ,y s ) as the starting point and (x t ,y t ) is the target point; A3, planning is completed when the node is expanded to the end point, and the set of child nodes is the current optimal path of the engineering truck.
4. The method for dynamic path planning of engineering truck group for intelligent construction site according to claim 1, wherein the step S3 is specifically: B1, designing a constraint tree, wherein the constraint tree comprises a constraint, a global path which is currently calculated and a global cost, and the description form of the constraint tree is (a i , v, t) which indicates that an engineering truck a i cannot occupy a grid v at a moment t; B2, detecting the conflict in the engineering truck single vehicle planning result calculated in the step S2, wherein the specific description form is (a i ,a j , v and t), namely that at the moment t, engineering trucks a i and a j occupy grids v at the same time, splitting a child node 1 and a child node 2 after the conflict is detected, distributing grids to a i in the child node 1, adding constraints (a i , v and t) to the nodes, distributing grids to a j in the child node 2, adding constraints (a j , v and t) to the nodes, applying the constraints of each child node to a floor planning algorithm, recalculating a path, then searching the node N with the minimum cost in the whole constraint tree, returning to the first step, continuing to search for the conflict, if the regenerated constraint tree has no conflict, repeating the process if the conflict exists.
5. The method for dynamic path planning of engineering truck group for intelligent construction site according to claim 1, wherein the step S4 is specifically: C1, according to the result obtained in the step S3, all inflection points in the path are sequentially extracted, two sets of rho= { p 1 |1< i < n } and a set delta= { p 1 } are created, wherein the set rho is used for representing all path nodes planned by an A-algorithm, p 1 and p n respectively represent a starting point and a target point, the set delta is used for storing a key path node, and the initial value is the starting point p 1 ; And C2, defining E (i, i+1) to represent the expansion direction between two adjacent nodes p i and p i+1 , if the E (i-1, i) and E (i, i+1) have the same direction, judging the node p i as a redundant node, otherwise, p i as a turning node, adding the turning node into the set delta, and using an expansion direction consistency rule to efficiently distinguish the nodes in the path so as to reduce the extra calculation amount.
6. The method for dynamic path planning of engineering truck group for intelligent construction site according to claim 1, wherein the step S5 specifically comprises: The distance measuring device can adopt an ultrasonic distance measuring device, the image acquisition device can use a camera, and the distance measuring direction at least comprises an upper direction, a lower direction, a left direction, a right direction, a front direction and a rear direction so as to obtain enough information, avoid dead angles and establish a rectangular coordinate system by utilizing three mutually perpendicular directions; And D2, calculating the increasing speed V 1 of the area of the object area in the continuous multi-frame images of the object by calculating the quotient of the area difference and the time difference of the area of the two adjacent images, and when the current distance from the edge of the engineering truck to the object is the minimum distance from the edge of the engineering truck to all objects in all ranging directions and the increasing speed V 0 of the area of the object area in the image with time is higher than a preset speed threshold.
7. The method for dynamic path planning of engineering truck group for intelligent construction site according to claim 1, wherein the step S6 is specifically: E1, if no collision risk exists, continuing to run according to the original route; If collision risk exists, evading measures are taken, and if the repulsive force potential field received by the engineering vehicle is U rep , the attractive force potential field is U att , the repulsive force is F rep and the attractive force is U att , the calculation formula of the attractive force potential field is as follows: Wherein m is a gravitational gain constant, m epsilon (0.01, 10), ρ is the distance between the target point and the engineering vehicle, and the gravitational force applied by the engineering vehicle is a negative gradient of gravitational potential field function: The repulsive force potential field U rep borne by the engineering truck is as follows: Wherein k is a repulsive force field constant, k epsilon (10, 100), q is the distance between the engineering vehicle and the obstacle, and gamma is the distance between the obstacle y and the engineering vehicle; and E3, adding a relative speed repulsive force into the repulsive force function, wherein the repulsive force potential field is as follows: wherein U v is the repulsive force potential field of relative speed, v 0 is the movement speed of the engineering truck relative to the obstacle, alpha is the included angle between v 0 and the x axis, the dynamic obstacle is simplified into a circle with radius r, r is the maximum size of the dynamic obstacle with collision risk, two tangent lines of the engineering truck to the obstacle are operated, and the included angles between the two tangent lines and the x axis are respectively The repulsive force F rep1 corresponding to the repulsive potential field is: e4, controlling the running track of the engineering vehicle according to the solution of the repulsive force and the attractive force obtained before; repulsive and attractive forces the resultant force calculation formula is: Wherein F att is the attractive force generated by the target point on the engineering vehicle, and F repi is the repulsive force generated by the ith barrier on the engineering vehicle.
8. The method for dynamic path planning of engineering truck group for intelligent construction site according to claim 1, wherein the step S7 is specifically: After the obstacle avoidance is completed, the engineering vehicle may deviate from the originally set path and needs to return to the original set path after the obstacle avoidance is completed, and the specific method is that the engineering vehicle and the dynamic obstacle are regarded as rigid bodies, the collision radiuses of the engineering vehicle and the dynamic obstacle are respectively R q and R o , a certain safety margin R a is reserved, and the safety distance R s between the engineering vehicle and the obstacle is defined: R s ＝R q +R o +R a The circumference with the center point of the obstacle as the center and the radius R s as the radius intersects with a certain point on the original planned track, which is the track recovery point, and the motion direction of the obstacle is perpendicular to the whole motion direction of the engineering truck, so that the selected recovery point is approximately symmetrical with the change point during obstacle avoidance, and the track is smoother.

Description

Engineering truck group dynamic path planning method for intelligent construction site Technical Field The invention relates to the technical field of engineering vehicles, in particular to an engineering vehicle group dynamic path planning method for an intelligent construction site. Background Along with the continuous promotion of the intelligent construction site concept, engineering vehicles are increasingly widely applied to construction sites. The engineering truck group (such as a transport truck, a concrete pump truck, a crane and the like) bears multiple tasks of material transportation, equipment transportation and the like in a construction site, so that the construction efficiency is greatly improved. However, the construction site has complex environment, limited space and dense distribution of personnel, mechanical equipment and barriers, and the engineering vehicle group faces many challenges such as path conflict, traffic jam, collision risk and the like in the operation process. At present, the dispatching and path planning of the construction site engineering vehicle depend on manual experience, and an intelligent and automatic path planning and obstacle avoidance means is lacked. The manual dispatching is not only low in efficiency, but also easy to cause vehicle running conflict due to information lag or judgment errors, and even causes safety accidents. Some construction sites have introduced a path planning system based on a static map, but in practical application, the construction site environment and the operation task are often changed, the static path planning is difficult to adapt to the dynamic environment, and the conditions such as sudden obstacles, temporary construction area adjustment and the like cannot be effectively treated. In addition, along with the expansion of the scale of construction sites and the increase of the number of engineering vehicles, the single vehicle path planning cannot meet the requirement of multi-vehicle collaborative operation. When engineering truck groups work simultaneously in a limited space, if an efficient dynamic path planning and coordination mechanism is lacked, vehicle congestion, resource waste and potential safety hazards are very easy to occur, and the overall construction progress and quality are influenced. Therefore, a method for planning a dynamic path of an engineering truck group, which is suitable for a dynamic environment of a construction site, supports multi-truck cooperation and has real-time obstacle avoidance capability, is needed. The method can dynamically adjust the running path of each engineering vehicle according to the real-time environment information of the construction site and the vehicle state, avoid collision and congestion, and improve the running efficiency of the vehicle and the safety of the construction site. The intelligent construction method has important significance in promoting construction of intelligent construction sites and realizing intelligent management of engineering vehicles. Disclosure of Invention Aiming at the problems that a worker Cheng Chequn in the existing intelligent construction site is low in path planning efficiency, path conflict and collision are easy to occur, real-time obstacle avoidance capability for dynamic obstacles is lacking and the like in a complex dynamic environment, the engineering vehicle group dynamic path planning method for the intelligent construction site is provided, a multi-node expansion A-based algorithm is adopted to generate a local optimal path for each engineering vehicle, then path conflicts among different engineering vehicles are detected and comprehensively solved through constraint trees, redundant nodes are removed through inflection point optimization strategies, the path length is shortened, smoothness is improved, a global optimal path is further generated, meanwhile, dynamic obstacle avoidance is carried out by combining with an artificial potential field method, the method can dynamically plan and adjust running paths of a plurality of engineering vehicles according to real-time environment information of the construction site, realize multi-vehicle cooperative obstacle avoidance, improve the intelligent and automatic level of path planning, and accordingly effectively improve the running efficiency of the engineering vehicle group and the safety of the construction site. The invention provides a dynamic path planning method for an engineering truck group for an intelligent construction site, which comprises the following steps of: The method comprises the steps of S1, modeling a space environment, dividing the searched space environment into grids with equal size, wherein each grid is called a unit, the position of the grid is determined by rows and columns of the unit, a search area is stored in a two-dimensional array, each item of the array represents one grid, the state space of the array is walkable and non-walkable, namely the gr