Unmanned aerial vehicle full-coverage path planning method fusing greedy strategy and A star algorithm

Abstract

The invention discloses an unmanned aerial vehicle full-coverage path planning method fusing a greedy strategy and an A star algorithm. The method belongs to the technical field of full-coverage path planning, and comprises the following operation steps: firstly, initializing a map and obstacles, and defining a marker quantity; secondly, evaluating attributes of surrounding points, and selecting a next point by using a greedy strategy; then, judging whether a dead zone is entered or not, and if so, re-planning a path to get rid of the dead zone by using an A star algorithm; if not, continuing to perform the greedy strategy to find the next advancing point; then, the coverage completion condition is checked, and whether each point is covered or not is checked; and finally, verifying the effectiveness of the full-coverage traversal algorithm in a three-dimensional environment. According to the invention, the problems of low efficiency, low coverage rate, high repetition rate, limitation in a complex environment and the like when the current technology is used for full-coverage traversal inspection, detection and cleaning of chemical industrial parks and coal mines and agricultural plant protection are solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of full-coverage path planning, and relates to the problem of full-coverage traversal path planning; specifically, it relates to a method for unmanned aerial vehicle full-coverage path planning that combines a greedy strategy and the A* algorithm (a method for unmanned aerial vehicle full-coverage path planning that combines an improved greedy strategy and the A* algorithm); it effectively solves the problem of covering the target area during the inspection process. Background Art

[0002] Currently, full-coverage traversal is a key function for robots, unmanned aerial vehicles, and other automated devices when performing complex tasks (such as environmental monitoring, coal mine inspection, cleaning, agricultural plant protection, etc.). Its goal is to ensure that every location within the task area is visited. The traditional A* algorithm is widely used in the field of path planning due to its high path planning efficiency and good optimization performance. However, in full-coverage tasks, its search space is large and it is prone to generating unnecessary repeated paths, making it difficult to meet the efficiency requirements in complex environments. At the same time, as a simple and efficient local search method, the greedy strategy can quickly select the current optimal solution, but due to the lack of a global perspective, it is easy to fall into a local optimum, resulting in some areas not being covered or path redundancy. In a complex dynamic environment, a single A algorithm or greedy strategy is difficult to balance global optimization and local real-time adjustment.

[0003] (Chinese Patent: Patent Number: CN106979785A): This method combines a bio-inspired neural network model and a backtracking mechanism, solves the deadlock problem of the robot through a dynamic A* algorithm, and at the same time introduces a market mechanism to optimize the selection of backtracking nodes, thereby reducing repeated coverage and total coverage time. Its drawback is that its adaptability to dynamic environments is weak. The backtracking mechanism mainly focuses on global deadlock problems, but the processing strategy for small dead zones in complex terrains is not clear enough, which may lead to a decrease in efficiency, does not pay attention to coverage uniformity, and complex algorithms may increase the demand for computing resources and have high requirements for hardware performance.

[0004] (Chinese Patents: Patent Numbers: CN118466301A and CN115727850A): The above patents all focus on the full-coverage path planning of multi-robot or multi-unmanned aerial vehicle systems, aiming to improve the coverage efficiency and reduce the task time-consuming through various methods. However, these methods still have certain limitations in specific scenarios. For example, when using the backtracking mechanism to solve the deadlock problem, the efficiency is low in scenarios with dense or sparse node distributions; the combination of the A* algorithm and reinforcement learning optimizes the monitoring path, but highly depends on the accuracy of the environmental model. Once the model is inaccurate, the monitoring efficiency will be affected; while the binary greedy strategy is easy to fall into a local optimum in complex environments and is difficult to achieve global optimal coverage, resulting in low efficiency. Summary of the Invention

[0005] In view of the above problems, the object of the present invention is to propose a method for path planning of full coverage of unmanned aerial vehicles (UAVs) that combines the greedy strategy and the A* algorithm (method for path planning of full coverage of UAVs that combines the greedy strategy and the A* algorithm).

[0006] The technical solution of the present invention is as follows: The method for path planning of full coverage of UAVs that combines the greedy strategy and the A* algorithm of the present invention is preferably a research method for path planning of full coverage traversal for gas inspection in chemical industrial parks and coal mines. The operation steps are as follows:

[0007] Step (1): Initialize the map and obstacles, and define flag variables.

[0008] Step (2): Calculate the information of the points around the current point and evaluate their attributes, add a mechanism for dynamically updating the coverage value, and improve the movement rules. Use the greedy strategy to select the next forward direction.

[0009] Step (3): Design a dead zone trigger mechanism. If any condition of the dead zone is reached, trigger an event and call the A* algorithm to re-plan a path to get out of the dead zone.

[0010] Step (4): Check the completion of coverage.

[0011] Step (5): Verify through three-dimensional environment simulation.

[0012] Further, the specific content of step (1) is as follows: Initialize the map and obstacles, and establish a grid map according to the layout of facilities in the chemical industrial park; and traverse each point on the map to determine whether the point is an obstacle. If it is an obstacle, mark it to indicate that it is not passable.

[0013] At the same time, set the starting point and path variables, and define a flag variable to detect whether entering the dead zone, that is, there are no accessible uncovered points.

[0014] Further, the specific content of step (2) is as follows: Obtain the coordinates of the 8 neighboring points around the current point and determine whether they are within the map boundary.

[0015] For each neighboring point within the boundary, evaluate its coverage attribute: If the coverage value is 0 and it is not an obstacle, mark it as a priority coverage point; if the coverage value is greater than 0 and it is not an obstacle, mark it as an accessed point; if it is an obstacle, mark it as a covered and non-passable point.

[0016] Calculate the heuristic value of each point, and select the point with the highest score as the next forward point; dynamically update the coverage value to avoid over-coverage.

[0017] Further, first, find all grid point indices within the current detection range of the UAV. Increase the point coverage value matrix T within the detection range by a step value r0, and determine whether the coverage value reaches the upper limit C. If it exceeds, limit it to C, that is, T(x,y) = min(T(x,y) + r0, C). Second, the UAV selects a point with the minimum coverage value within the coverage range as the traction point. The traction point is the target point of the UAV's movement, and by comparing with the direction angle of the current UAV, the nearest target point is selected. This method enables the UAV to select the area that most needs to be covered instead of moving blindly, where the angle difference between the UAV and the traction point is θ i = arctan(d yi ,d xi );

[0018] In case of special situations, call the improved movement rule to solve the path planning selection problem in specific scenarios; first, judge the information of surrounding points. The around_points array contains the states of each point around the current position, and the movement direction is determined by judging whether these points are empty; second, assume the current point P0(x0,y0), and the surrounding points are P1, P2, P3, P4, P5, P6, P7, P8 respectively. The state of each point can be expressed as:

[0019]

[0020] Finally, when running to the end of the path or a T-junction, select the dead zone with the minimum relative path, that is, when P1 = 1, P4 = 1, P6 = 1, P2 = 0, P7 = 0, call the improved movement rule, and select the next forward point by judging the position of the current point P0(x0,y0), that is:

[0021]

[0022] At the same time, according to the innovative mechanism of dynamically updating the coverage value, the coverage value of the priority coverage points is adjusted in real time to ensure that the coverage value accumulates gradually but does not exceed the set limit value, so as to optimize the path selection and improve the coverage efficiency.

[0023] Further, the specific content of step (3) is: design a dead zone trigger mechanism to judge whether it enters the dead zone. If all surrounding points have been covered or are obstacles, trigger an event, input the trapped information, and plan a new path to get out of the dead zone by combining and calling the A* algorithm, and start traversing again.

[0024] Further, the specific content of step (4) is: check all points in the map to judge whether they have all been covered. If not, perform step (2). If full coverage is completed, end the path planning.

[0025] Further, the specific steps of step (5) are as follows: First, perform 3D modeling of the chemical industrial park in Blender. Second, import the modeled 3D model into the gazebo environment. Finally, perform traversal simulation in the 3D environment.

[0026] The beneficial effects of the present invention are as follows: 1. It makes up for the lack of flexibility and poor adaptability in path planning of traditional methods, can achieve fast coverage in a simple environment, and can also dynamically adjust the path in a complex environment to ensure the integrity of coverage; 2. Adopt the ladder method to dynamically update the coverage values of grid points within the detection range and limit the maximum coverage value, solve the resource waste caused by uncontrolled coverage values in traditional algorithms, and ensure coverage efficiency and coverage uniformity; 3. By improving the movement rules, preferentially turn into dead zones with smaller paths, effectively reduce duplicate paths, and improve the efficiency of full-coverage traversal. It specifically targets complex terrains such as T-junctions, optimizes path selection, and ensures coverage rate and flexibility. This strategy can also adapt to boundary conditions, avoid omissions or redundancies, and improve the intelligence and adaptability of the overall algorithm; 4. Design a dead zone trigger mechanism. If all the surrounding points of the current point meet the trigger conditions, the event will be immediately triggered and the A* algorithm will be called to avoid the drone wasting time in this area and improve the coverage efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 is the flowchart of the full-coverage traversal path planning research method of the present invention;

[0028] Figure 2 is the flowchart of the full-coverage traversal path planning algorithm of the present invention;

[0029] Figure 3 is the effect diagram of the full-coverage traversal path planning of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0030] The following further elaborates on the specific technical solutions of the present invention in combination with specific examples.

[0031] As shown in the figure, the present invention relates to an unmanned aerial vehicle full-coverage path planning method that combines a greedy strategy and the A* algorithm;

[0032] The operation steps are as follows: First, initialize the map and obstacles, and define flag variables;

[0033] Second, evaluate the attributes of the surrounding points and use the greedy strategy to select the next point;

[0034] After that, determine whether to enter a dead zone. If entering, use the A* algorithm to re-plan a path to get out of the dead zone. If not entering the dead zone, continue to use the greedy strategy to find the next forward point;

[0035] Finally, check the coverage completion situation and check whether each point is covered.

[0036] The specific implementation steps are as follows:

[0037] S101: As shown in Figure (2), it is the flowchart of the full-coverage path planning algorithm. First, initialize the map and obstacles, and establish a grid map; traverse each point on the map and determine whether the point is an obstacle. If it is an obstacle, mark it as covered, indicating that it is impassable;

[0038] S102: Define the starting point and path variables, and define a flag to determine whether to enter a dead end, that is, all surrounding points have been covered or are obstacles;

[0039] S201: Calculate the coordinates of the 8 neighboring points in the surrounding directions of the current point, and determine whether they are within the map boundary;

[0040] S202: Evaluate the coverage attributes of the surrounding points: If the coverage value is 0 and it is not an obstacle, mark it as a priority coverage point; if the coverage value is greater than 0 and it is not an obstacle, mark it as an accessed point; if it is an obstacle, mark it as a covered and impassable point;

[0041] Calculate the heuristic value of each point, and select the point with the highest score as the next advancing point;

[0042] In case of special situations, improve the movement rules to solve the path planning selection problem in specific scenarios. First, judge the information of the surrounding points. The around_points array contains the states of each point around the current position. Determine the movement direction by judging whether these points are empty. Secondly, assume the current point P0(x0,y0), and the surrounding points are P1, P2, P3, P4, P5, P6, P7, P8 respectively. The state of each point can be expressed as:

[0043]

[0044] Finally, when running to the end of the path or a T-junction, select the dead end with the smallest relative path, that is, when P1 = 1, P4 = 1, P6 = 1, P2 = 0, P7 = 0, call the improved movement rule, and select the next advancing point by judging the position of the current point P0(x0,y0), that is

[0045]

[0046] S203: Incorporate the method of dynamically updating the coverage value. First, find all the grid point indices within the detection range of the current UAV. Increase the coverage value matrix T of the points within the detection range by a step value r0, and determine whether the coverage value reaches the upper limit C. If it exceeds, limit it to C, that is, T(x,y) = min(T(x,y) + r0, C). Second, the UAV selects a point with the minimum coverage value within the coverage range as the traction point. The traction point is the target point of the UAV's movement, and by comparing with the direction angle of the current UAV, the nearest target point is selected. This method enables the UAV to select the area that most needs to be covered instead of moving blindly. The angle difference between the UAV and the traction point is θ i = arctan(d yi ,d xi );

[0047] S301: Design a dead zone judgment mechanism. If all the points around the current point are satisfied as being covered or being obstacles, it is determined that the UAV has entered the dead zone and an event is triggered.;

[0048] S302: When it is determined that the current point has fallen into the dead zone, as shown in Figure (3), after an event is triggered when a certain point falls into the dead zone, the A* algorithm is called for path planning. The next optimal point is selected through f = g + h, rather than simply selecting the point with the shortest straight-line distance. That is, find a path from the current point to the nearest uncovered area, jump out of the dead zone, and continue to use the greedy strategy to select the next forward point;

[0049] S401: Check whether each point in the map has been covered. If all have been covered, end the path planning.

[0050] As Figure 3 shown, where (a) is the effect diagram of the A* round-trip full-coverage traversal path planning. The results show that the path length is 740, the number of turning times is 113, and the repetition rate is as high as 50%; (b) is the effect diagram of the full-coverage traversal path planning integrating the greedy strategy and the A* algorithm. The results show that the path length is 366.8, the number of turning times is 94, and the repetition rate is 3.6%; (c) is the effect diagram of the path planning further improved on the basis of integrating the greedy strategy and the A* algorithm. The results show that the path length is 366.2, the number of turning times is 88, and the repetition rate is only 3.4%. Through the result comparison, it shows that the integrated improved algorithm described in the present invention has greatly improved in terms of path length, number of turning times, and repetition rate, effectively solving the problems of many turning times, long path, and high repetition rate in the process of full-coverage traversal by the conventional A* round-trip algorithm;

[0051] S501: First, perform 3D environmental modeling of the chemical industrial park in Blender. Secondly, import the modeled 3D model into Gazebo. After that, turn on the quadcopter drone in the imported 3D chemical industrial park. Finally, use the improved fusion algorithm to conduct a full-coverage traversal simulation experiment. The final simulation experiment shows that the drone can perform full-coverage traversal flight inspections in a 3D environment.

[0052] The present invention solves the problems of low efficiency, low coverage rate, high repetition rate, and limitations in complex environments in the current technology for full-coverage traversal inspection, detection, cleaning, and agricultural plant protection in chemical industrial parks and underground coal mines. First, by integrating the greedy strategy and the A* algorithm, combining the local advantages of the greedy strategy and the global advantages of the A* algorithm, it makes up for the lack of flexibility and poor adaptability in path planning of traditional methods. It can achieve fast coverage in a simple environment and dynamically adjust the path in a complex environment to ensure the integrity of coverage. Secondly, a method for dynamically updating the coverage value is added to avoid a decrease in efficiency caused by over-coverage. After that, the movement rules are improved to solve the movement in special situations such as T-junctions and the ends of the map boundary, reduce the repeated path, and improve the efficiency at the same time. Finally, a dead zone trigger mechanism is designed. If the current point falls into the dead zone, an event is triggered to quickly call the A* algorithm to get out of trouble and get rid of the local optimum situation. It effectively improves the efficiency of full-coverage.

Claims

1. A method for unmanned aerial vehicle full-coverage path planning that integrates a greedy strategy and the A-star algorithm, characterized in that, The operation steps are as follows: Step (1): Initialize the map and obstacles, and define flag variables. Step (2): Calculate the information of the points around the current point and evaluate their attributes, add a mechanism for dynamically updating the coverage value, and improve the movement rules. Use the greedy strategy to select the next forward direction. Step (3): Design a dead zone trigger mechanism. If any condition of the dead zone is reached, trigger an event and call the A* algorithm to re-plan a path to get out of the dead zone. Step (4): Check the completion of coverage. Step (5): Conduct three-dimensional environment simulation verification.

2. The method for full coverage path planning of an unmanned aerial vehicle by integrating a greedy strategy and an A-star algorithm according to claim 1, wherein Specifically, in step (1): Initialize the map and obstacles, and establish a grid map. And traverse each point on the map to determine whether the point is an obstacle. If it is an obstacle, mark it to indicate that it is impassable. At the same time, set the starting point and path variables, and define a flag variable to detect whether entering the dead zone, that is, there are no accessible uncovered points.

3. The method for unmanned aerial vehicle full-coverage path planning integrating the greedy strategy and the A* algorithm according to claim 1, characterized in that, Specifically, in step (2): Obtain the coordinates of the 8 neighboring points around the current point and determine whether they are within the map boundary. For each neighboring point within the boundary, evaluate its coverage attribute: If the coverage value is 0 and it is not an obstacle, mark it as a priority coverage point; if the coverage value is greater than 0 and it is not an obstacle, mark it as an accessed point; if it is an obstacle, mark it as a covered and impassable point. Calculate the heuristic value of each point and select the point with the highest score as the next forward point. Dynamically update the coverage value to avoid over-coverage.

4. The UAV full-coverage path planning method integrating the greedy strategy and the A* algorithm according to claim 3, characterized in that Specifically, first, find all grid point indices within the current UAV detection range, increase the coverage value matrix T of the points within the detection range by a step value r0, and determine whether the coverage value reaches the upper limit C. If it exceeds, limit it to C, that is, T(x,y) = min(T(x,y) + r0, C). Secondly, the UAV selects a point with the smallest coverage value within the coverage range as the traction point. The traction point is the target point of the UAV's movement, and by comparing with the direction angle of the current UAV, select the nearest target point. The above method enables the drone to select the area most in need of coverage, where the angle difference between the drone and the towing point is θ i = arctan(d yi , d xi ).

5. The method for full coverage path planning of an unmanned aerial vehicle that combines a greedy strategy and the A* algorithm according to claim 4, wherein In case of special situations, call the improved movement rules to solve the path planning selection problem in specific scenarios. First, judge the information of the surrounding points. The around_points array contains the states of each point around the current position. Determine the movement direction by judging whether these points are empty. Secondly, assume the current point P0(x0,y0), and the surrounding points are P1, P2, P3, P4, P5, P6, P7, P8 respectively. The state of each point is expressed as: Finally, when running to the end of the path or a T-junction, select the dead zone with the smallest relative path, that is, when P1 = 1, P4 = 1, P6 = 1, P2 = 0, P7 = 0, call the improved movement rules, and select the next forward point by judging the position of the current point P0(x0,y0), that is: At the same time, according to the innovative mechanism of dynamically updating the coverage value, adjust the coverage value of the priority coverage points in real time to ensure that the coverage value accumulates gradually but does not exceed the set limit value, so as to optimize the path selection and improve the coverage efficiency.

6. The method for full-coverage path planning of an unmanned aerial vehicle that combines a greedy strategy and the A* algorithm according to claim 1, wherein The specific content of step (3) is as follows: Design a dead zone trigger mechanism to determine whether the dead zone is entered. If all surrounding points have been covered or are obstacles, trigger an event, input the trapped information, plan a new path to get out of the dead zone by combining and calling the A* algorithm, and start traversing again.

7. The method for full coverage path planning of an unmanned aerial vehicle by integrating a greedy strategy and the A-star algorithm according to claim 1, wherein The specific content of step (4) is as follows: Check all points in the map to determine whether they have all been covered. If not, perform step (2). If full coverage is completed, end the path planning.

8. The method for unmanned aerial vehicle full-coverage path planning integrating the greedy strategy and the A* algorithm according to claim 1, wherein The specific content of step (5) is as follows: Import the three-dimensional model of the chemical plant built in Blender into Gazebo and perform full-coverage traversal simulation in its environment to verify the effectiveness of the algorithm.

Images