Underwater robot path planning method for obstacle avoidance in complex underwater environment

Abstract

The invention provides an underwater robot path planning method for obstacle avoidance in a complex underwater environment, and the method comprises the steps: 1, determining a search and rescue task region, marking static obstacles in the search and rescue task region, and building a grid map; 2, acquiring an initial obstacle avoidance path of a static obstacle according to an improved ISSA algorithm; 3, the AUV executes path tracking according to the initial obstacle avoidance path, updates dynamic obstacle information detected by the forward-looking sonar in real time to a grid map, and starts an improved IDWA algorithm to obtain a dynamic obstacle avoidance path; and 4, after the AUV arrives at the target point, detecting a target object through a sensor, uploading detection information to a shore base end, judging whether a search and rescue task is completed or not, and if not, continuing to execute the steps 2-3 until the search and rescue task is completed. According to the invention, the safety of the underwater robot in the process of executing the underwater search and rescue task can be improved, the detection task of the search and rescue target object is efficiently realized, and the efficiency of the underwater search and rescue task is effectively improved.

Description

Technical Field

[0001] The invention relates to an underwater robot path planning method for avoiding obstacles in a complex underwater environment, and belongs to the technical field of underwater robots. Background Art

[0002] With the recovery of global trade and the increase in cargo transportation in waters, people have found that shipwrecks have brought great harm to the safety of life and property of crew members. It is necessary to improve the efficiency of search and rescue missions after shipwrecks in the hope of effectively protecting the safety of people's lives and property. In shipwreck search and rescue missions, new mission characteristics are usually shown, such as a large search and rescue range, a high risk factor, and a high timeliness of search and rescue missions. This has brought huge technical challenges to the traditional solution of relying on divers to carry out search and rescue missions.

[0003] In recent years, with the development of technologies such as intelligent control, navigation and communication, underwater detection, and planning and decision-making, the application of various technologies in underwater robots has become increasingly mature, which has laid a technical foundation for underwater robots in detecting underwater environmental information and determining the location and identity of people in distress.

[0004] At present, most underwater robots plan their mission paths based on prior knowledge, and do not fully consider the changes in unknown environmental information, making unknown obstacles a greater threat to the safety of underwater robots. On the one hand, the preset path limits the performance of underwater robots. It makes it impossible for underwater robots to make real-time adjustments based on environmental and mission information, and it is impossible to meet the timeliness requirements of search and rescue missions; on the other hand, traditional solutions are difficult to deal with the threat of dynamic obstacles, so that underwater robots cannot respond to dynamic obstacles in time, resulting in mission failure. Based on the needs of underwater search and rescue missions, the present invention proposes a path planning method for underwater robots for obstacle avoidance in complex underwater environments. Summary of the invention

[0005] The present invention aims to solve the technical problem that an underwater robot cannot make real-time adjustments according to the environment and mission information and cannot respond to dynamic obstacles in time, and further proposes an underwater robot path planning method for avoiding obstacles in complex underwater environments.

[0006] The technical solution adopted by the present invention to solve the above problems is: the present invention proposes a path planning method for an underwater robot for avoiding obstacles in a complex underwater environment, comprising:

[0007] Step 1: Determine the search and rescue mission area, mark the static obstacles in the search and rescue mission area and create a grid map;

[0008] Step 2: Obtain the initial obstacle avoidance path of the static obstacle according to the improved ISSA algorithm, send the initial obstacle avoidance path to the path planning module, track the path points as local nodes of the IDWA algorithm, and complete the AUV's static obstacle avoidance;

[0009] Step 3: The AUV performs path tracking according to the initial obstacle avoidance path, and updates the dynamic obstacle information detected by the forward-looking sonar in real time to the grid map, starts the improved IDWA algorithm to obtain the dynamic obstacle avoidance path, and passes the dynamic obstacle avoidance path to the path tracking module to complete the AUV's dynamic obstacle avoidance;

[0010] Step 4: After the AUV reaches the target point, it detects the target object through the sensor, uploads the detection information to the shore base, and determines whether the search and rescue mission is completed. If not, continue to execute steps 2-3 until the search and rescue mission is completed.

[0011] Optionally, during the movement of the AUV in step 2 and step 3, the position of the obstacle and the distance from the AUV body are sensed by the forward-looking sonar, the obstacle is expanded into a circular obstacle, the AUV is expanded into a circular safety area, and the AUV circular safety area is added to the circular obstacle radius, and the AUV is simplified into a mass point;

[0012] The expression for the radius of a circular obstacle is:

[0013] R b =R auv +R ob (1);

[0014] In formula (1), R auv is the radius of AUV after expansion, R ob is the actual radius of the obstacle, R b is the radius used in the simulation.

[0015] Optionally, the steps of completing the static obstacle avoidance of the AUV in step 2 include:

[0016] Step 2.1: Determine the AUV starting point and static obstacle position;

[0017] Step 2.2: Initialize the population position and place the sparrows at the initialization position to start exploring;

[0018] Step 2.3: Calculate the fitness of each sparrow and obtain the global optimal and worst positions;

[0019] Step 2.4: Update individual optimal position, global optimal position and fitness;

[0020] Step 2.5: Update the global optimal path based on the individual optimal position, the global optimal position and fitness;

[0021] Step 2.6: Determine whether the maximum number of iterations is met. If not, repeat steps 2.3 to 2.5 until the maximum number of iterations is met.

[0022] Step 2.7: Output the optimal static obstacle avoidance path for the AUV.

[0023] Optionally, in step 2.4, updating the individual optimal position, the global optimal position and the fitness specifically includes:

[0024] Step 2.4.1: Introduce the dynamic weight coefficient γ into the discoverer formula to obtain the discoverer dynamic weight coefficient;

[0025] Step 2.4.2: Introduce the finder ratio coefficient η, and obtain the improved finder position based on the finder dynamic weight coefficient and the finder ratio coefficient η;

[0026] Step 2.4.3: Obtain the positions of known static obstacles in the grid map during the exploration process based on the improved discoverer position;

[0027] Step 2.4.4: Remove the way of approaching the origin during the convergence process of the participant position formula, and only retain the information of the participant approaching the optimal fitness function position, obtain the improved participant position, and obtain the position of the unknown static obstacle in the grid map based on the improved participant position;

[0028] Step 2.4.5: Update the position of the AUV in the grid map in real time, and update the individual optimal position, the global optimal position and the fitness based on the position of the AUV in the grid map, the positions of the known static obstacles in the grid map and the positions of the unknown static obstacles in the grid map;

[0029] The expression of the dynamic weight coefficient γ is:

[0030]

[0031] The expression of the finder's proportional coefficient η is:

[0032]

[0033] The improved calculation formula for the finder's position is:

[0034]

[0035] The improved joiner position formula is:

[0036]

[0037] Optionally, the steps of completing the AUV's dynamic obstacle avoidance in step 3 include:

[0038] Step 3.1: Initialize the grid map and IDWA algorithm parameters;

[0039] Step 3.2: Receive dynamic obstacle information detected by the forward-looking sonar and update the initialized grid map;

[0040] Step 3.3: Sample the velocity of the AUV and generate the trajectory space;

[0041] Step 3.4: Introduce the threshold D related to the AUV's own performance auv Calculate the AUV dynamic weight and update the AUV's motion trajectory;

[0042] Step 3.5: Use the improved DWA algorithm evaluation function to evaluate the motion trajectory of the updated AUV;

[0043] Step 3.6: Select the best trajectory among the trajectories of the objects that have not collided with the obstacles according to the evaluation results, and execute the best path;

[0044] AUV's own performance related threshold D auv The calculation formula is:

[0045]

[0046] In formula (6), ξ is a fixed coefficient value used to adjust the relationship between the AUV scale and its speed, D auv The maximum linear velocity u of the AUV max Proportional to linear acceleration Inversely proportional.

[0047] Optionally, the step of AUV dynamic weight in step 3.4 includes: dynamic speed weight α v , Dynamic heading weight coefficient b v and dynamic distance weight coefficient c v .

[0048] Optionally, the step of calculating the AUV dynamic weight in step 3.4 includes:

[0049] Step 3.4.1: Based on the AUV's own performance-related threshold D auv and the distance D between the dynamic obstacle and the AUV min Calculate the dynamic speed weight α v ;

[0050] Step 3.4.2: By comparing the AUV's own performance related threshold D auv and the distance D between the dynamic obstacle and the AUV min Design dynamic heading weight coefficient b v ;

[0051] Step 3.4.3: Distance D from the AUV through the dynamic obstacle min Design dynamic distance weight coefficient c v ;

[0052] Dynamic speed weight α v The expression is:

[0053]

[0054] Dynamic heading weight coefficient b v The expression is:

[0055]

[0056] Evaluation index b closely related to heading angle σ The calculation formula is:

[0057]

[0058] ρ = -1·(θ-90)(10);

[0059] In formulas (9) and (10), b v is the heading dynamic weight coefficient, which is related to the AUV heading information; b σ is an evaluation index closely related to the heading angle. ρ is related to the heading angle of the AUV and is used to distinguish the AUV orientation. θ is the heading angle in the AUV's own coordinate system. max and b min is the heading coefficient of AUV under extreme conditions;

[0060] Dynamic distance weight coefficient c v The calculation formula is:

[0061]

[0062] Optionally, the expression of the improved DWA algorithm evaluation function in step 3.5 is:

[0063] G(ν,ω)=a ν velocity(ν,ω)+b ν ·geading(ν,ω)+c v ·dist(v,ω) (12);

[0064] In formula (12), ω is the angular velocity of the robot, velocity(ν,ω) is the velocity evaluation function, heading(ν,ω) is the heading evaluation function, and dist(v,ω) is the distance evaluation function.

[0065] The beneficial effects of the present invention are:

[0066] 1. The present invention can improve the safety of underwater robots in the process of performing underwater search and rescue tasks, and can efficiently realize the detection task of search and rescue target objects, effectively improving the efficiency of underwater search and rescue tasks.

[0067] 2. The present invention divides the underwater obstacle avoidance task into two steps, which can give full play to the performance characteristics and advantages of different sensors of the underwater robot, and can help the AUV to autonomously and efficiently complete the obstacle avoidance task in complex mission scenarios of search and rescue missions, effectively ensuring the safety of the AUV in performing the task. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 A flow chart of a method for underwater robot path planning for avoiding obstacles in complex underwater environments provided by the present invention;

[0069] Figure 2 A flow chart of the improved sparrow algorithm provided by the present invention;

[0070] Figure 3 A flowchart of the improved DWA algorithm provided by the present invention;

[0071] Figure 4 A schematic diagram of the obstacle expansion process provided by the present invention;

[0072] Figure 5 A schematic diagram of an underwater robot obstacle avoidance solution provided by the present invention;

[0073] Figure 6 The ISSA-IDWA fusion algorithm provided by the present invention obtains the optimal path schematic diagram;

[0074] Figure 7 The dynamic weight change curve diagram provided by the present invention;

[0075] Figure 8 This is a curve diagram of the change of the finder's proportional coefficient provided by the present invention. DETAILED DESCRIPTION

[0076] Combination Figure 1-8 This embodiment is described as follows. Figure 1 As shown, the steps of a method for underwater robot path planning for avoiding obstacles in a complex underwater environment described in this embodiment include:

[0077] S1: Determine the search and rescue mission area through known information and expert system, and construct a grid map of the known static obstacle locations such as islands and coral reefs;

[0078] When AUV performs underwater search and rescue missions, obstacles in unknown environments are mainly divided into two categories: dynamic and static. Obstacle avoidance research will be carried out on these two types of obstacles respectively:

[0079] (1) Static obstacles

[0080] Most obstacles that appear in unknown environments are static obstacles. This type of obstacle mainly includes underwater shipwreck structures, underwater reefs and artificial structures, which are widely distributed and vary greatly in size. Usually, this type of obstacle is expanded according to the method in Chapter 3 and simplified into a square obstacle in the grid map. The location distribution of static obstacles in the environment can be obtained in advance by executing the sensor information in the global coverage path.

[0081] (2) Dynamic obstacles

[0082] During underwater search and rescue, a special type of obstacle will be encountered - dynamic obstacles. Dynamic obstacles are usually floating objects that drift with the water flow, schools of fish in the water, etc. Traditional obstacle avoidance solutions are difficult to achieve safe obstacle avoidance. It is necessary to predict their position information in real time through sensor information and avoid obstacles in real time. In order to ensure the obstacle avoidance effect of dynamic obstacles, dynamic obstacles are usually expanded into circular obstacles, which can ensure that the AUV can completely avoid them in all directions and fully guarantee the safety of the AUV.

[0083] During the movement of AUV, the forward-looking sonar can be used to sense the location of obstacles and their distance from the AUV. In order to simplify the calculation and ensure the safety redundancy of AUV, Figure 4 As shown in FIG. 1 , this embodiment simplifies the robot and the obstacle. In order to facilitate calculation and simulation, the moving obstacle is expanded into a circular obstacle. In addition, in order to ensure the safety of the AUV, the AUV itself is also expanded into a circular safety area, and the simplified AUV circular area radius is added to the obstacle radius. The process is as follows: Figure 3 The final obstacle expansion radius is determined as follows:

[0084] R b =R auv +R ob (1);

[0085] In formula (1), R auv is the radius of AUV after expansion, R ob is the actual radius of the obstacle, R b is the radius used in the simulation.

[0086] S2: Obtain the obstacle avoidance path of the static obstacle based on the improved ISSA algorithm and send it to the path planning module. Track its path points as local nodes of the IDWA algorithm to ensure the global optimality of the obstacle avoidance path.

[0087] The planner on the AUV obtains the obstacle avoidance path of static obstacles based on the ISSA algorithm, and tracks its path points as local nodes of the IDWA algorithm to ensure the global optimality of the obstacle avoidance path. This implementation aims to solve the problem that the traditional SSA algorithm falls into the local optimal solution in the initial stage and cannot determine the optimal solution at the end of the algorithm. By introducing the dynamic weight coefficient and the discoverer ratio coefficient, and adjusting the two variables, the number of discoverers in the initial stage is large and the weight is high; the number of joiners in the final stage is large and the weight is high. This can effectively improve the overall optimization ability and efficiency of the sparrow algorithm, such as Figure 2 As shown, the steps of the improved sparrow algorithm are:

[0088] S201: Determine the AUV starting point and the static obstacle position;

[0089] S202: Initialize the population position, place the sparrows at the initialization position and start exploring;

[0090] S203: Calculate the fitness of each sparrow and obtain the global optimal and worst positions;

[0091] S204: Figure 7 As shown, this embodiment introduces a dynamic weight coefficient γ into the discoverer formula. By analyzing formula (2), it can be seen that the γ value gradually decreases as t increases. In the algorithm, it is shown that the weight of the discoverer in the early stage is relatively large, which can effectively improve the population search range and increase the richness of the population solution. In the later stage of the algorithm, the γ value becomes smaller, the weight factor decreases, the global search ability weakens, and the local optimization ability is strengthened, so that the algorithm converges quickly. The dynamic weight coefficient γ of the discoverer is as follows:

[0092]

[0093] S205: Figure 8 As shown, this embodiment introduces the discoverer ratio coefficient η. By analyzing formula (3), it can be seen that as t increases, the proportion of discoverers in the population decreases, and the proportion of joiners increases. In the early stage of the algorithm, a large-scale search can be achieved. In the later stage, by increasing the number of joiners, the overall search efficiency of the local optimum can be improved. The discoverer ratio coefficient η is as follows:

[0094]

[0095] The improved finder position can be expressed as:

[0096]

[0097] S206: When i>n / 2, the joiner position formula is the product of a normally distributed random number and an exponential function with the natural logarithm as the base, which can converge to 0, and the joiner will move closer to the origin. In order to obtain the optimal solution faster, this implementation deletes the way of moving closer to the origin, and only retains the information of the joiner moving closer to the optimal fitness function position. The improved joiner position formula can be expressed as:

[0098]

[0099] S3: The AUV performs path tracking according to the initial obstacle avoidance path, and updates the dynamic obstacle information detected by the sensor to the grid map, starts the improved IDWA algorithm to obtain the dynamic obstacle avoidance path, and passes the dynamic obstacle avoidance path to the path tracking module, so that the AUV can complete the dynamic obstacle avoidance, such as Figure 3 As shown, the steps to improve the DWA algorithm include:

[0100] S301: Initialize the grid map and IDWA algorithm parameters;

[0101] S302: receiving dynamic obstacle information detected by the forward-looking sonar and updating the initialized grid map;

[0102] S303: Sampling the velocity of the AUV to generate a trajectory space;

[0103] S304: Introducing AUV's own performance-related threshold D auv Calculate the AUV dynamic weight and update the AUV's motion trajectory;

[0104] AUV's own performance related threshold D auv The calculation formula is:

[0105]

[0106] In formula (6), ξ is a fixed coefficient value, which is used to adjust the relationship between the AUV scale and its speed. Since the AUV operates in a complex underwater environment, its value is set to 10. auv The maximum linear velocity u of the AUV max Proportional to linear acceleration Inversely proportional.

[0107] S30401: In the DWA algorithm, the weight of the speed term is related to the distance D of the obstacle from the AUV. min The speed term weight is closely related to the speed of the AUV. When the distance between the AUV and the obstacle is small, the speed term weight decreases. A smaller speed term weight can ensure the safety of the AUV when it is close to the obstacle. When the AUV is far away from the obstacle, a larger speed term is required to ensure the global efficiency of the AUV. The dynamic speed weight a designed in this embodiment v , as shown below:

[0108]

[0109] S30402: After completing the above adaptive speed item weight design, in order to better ensure the safety of the AUV, the heading weight of the DWA algorithm is improved. When the AUV performs dynamic obstacle avoidance, it will encounter dynamic obstacles from all sides. The threat in the heading direction mainly comes from obstacles within 180° of the AUV's direction of travel. When the AUV faces an obstacle, the heading options are left turn and right turn. Designing a suitable heading weight is the key to ensure the effective obstacle avoidance of the AUV. When the distance between the AUV and the obstacle is less than D min , at this time the dynamic weight coefficient b v Will follow D min The dynamic weight coefficient b designed in this embodiment decreases and gradually increases. v , as shown below:

[0110]

[0111] Evaluation index b closely related to heading angle σ The calculation formula is:

[0112]

[0113] ρ = -1·(θ-90)(10);

[0114] In formulas (9) and (10), b v is the heading dynamic weight coefficient, which is related to the AUV heading information; b σ is an evaluation index closely related to the heading angle. ρ is related to the heading angle of the AUV and is used to distinguish the AUV orientation. θ is the heading angle in the AUV's own coordinate system. max and b min is the heading coefficient of AUV under extreme conditions;

[0115] S30403: The distance weight between the AUV and the obstacle plays a decisive role in whether the AUV can achieve dynamic obstacle avoidance. When the distance between the AUV and the obstacle is close, a larger distance weight is required to ensure the safety of the AUV. When the AUV is far away from the obstacle, the distance weight needs to be reduced so that the AUV can quickly reach the target point. The dynamic weight coefficient c designed in this implementation is v , as shown below:

[0116]

[0117] S305: Evaluate the motion trajectory of the updated AUV using the improved DWA algorithm evaluation function;

[0118] The expression of the improved DWA algorithm evaluation function is:

[0119] G(v,ω)=a v velocity(ν,ω)+b v heading(v,ω)+c v ·dist(v,ω) (12);

[0120] In formula (12), ω is the angular velocity of the robot, velocity(v,ω) is the velocity evaluation function, heading(v,ω) is the heading evaluation function, and dist(v,ω) is the distance evaluation function.

[0121] S306: Figure 6 As shown, the optimal trajectory among the motion trajectories of the non-colliding obstacles is selected according to the evaluation results, and the optimal path is executed.

[0122] S4: Figure 5 As shown, after executing the optimal path to make the AUV arrive at the target point, the sensor completes the detection of the target object, and uploads the detection information to the shore-based end, and determines whether the search and rescue mission has been completed. If not, continue to execute S2-S3 until the search and rescue mission is completed.

[0123] This implementation method divides the underwater obstacle avoidance task into two steps, which can give full play to the performance characteristics and advantages of different sensors of the underwater robot, and can help the AUV to autonomously and efficiently complete the obstacle avoidance task in complex mission scenarios of search and rescue missions, effectively ensuring the safety of the AUV in performing the mission.

[0124] The above is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technician familiar with this profession can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent replacement and improvement made to the above embodiments without departing from the content of the technical solution of the present invention, based on the technical essence of the present invention, within the spirit and principles of the present invention, still fall within the protection scope of the technical solution of the present invention.

Claims

1. A method for underwater robot path planning for obstacle avoidance in complex underwater environments, characterized in that: The steps of the underwater robot path planning method for avoiding obstacles in a complex underwater environment include: Step 1: Determine the search and rescue mission area, mark the static obstacles in the search and rescue mission area and create a grid map; Step 2: Obtain the initial obstacle avoidance path of the static obstacle according to the improved ISSA algorithm, send the initial obstacle avoidance path to the path planning module, track the path points as local nodes of the IDWA algorithm, and complete the AUV's static obstacle avoidance; Step 3: The AUV performs path tracking according to the initial obstacle avoidance path, and updates the dynamic obstacle information detected by the forward-looking sonar in real time to the grid map, starts the improved IDWA algorithm to obtain the dynamic obstacle avoidance path, and passes the dynamic obstacle avoidance path to the path tracking module to complete the AUV's dynamic obstacle avoidance; Step 4: After the AUV reaches the target point, it detects the target object through the sensor, uploads the detection information to the shore base, and determines whether the search and rescue mission is completed. If not, continue to execute steps 2-3 until the search and rescue mission is completed.

2. The underwater robot path planning method for avoiding obstacles in complex underwater environments according to claim 1, characterized in that: During the movement of the AUV in steps 2 and 3, the forward-looking sonar is used to sense the position of the obstacle and the distance from the AUV body, and the obstacle is expanded into a circular obstacle, the AUV is expanded into a circular safety area, and the AUV circular safety area is added to the radius of the circular obstacle, simplifying the AUV into a particle; The expression for the radius of a circular obstacle is: R b =R auv +R ob (1); In formula (1), R auv is the radius of AUV after expansion, R ob is the actual radius of the obstacle, R b is the radius used in the simulation.

3. The underwater robot path planning method for avoiding obstacles in complex underwater environments according to claim 1 is characterized in that: The steps to complete the AUV's static obstacle avoidance in step 2 include: Step 2.1: Determine the AUV starting point and static obstacle position; Step 2.2: Initialize the population position, place the sparrows at the initialization position and start exploring; Step 2.3: Calculate the fitness of each sparrow and obtain the global optimal and worst positions; Step 2.4: Update individual optimal position, global optimal position and fitness; Step 2.5: Update the global optimal path based on the individual optimal position, the global optimal position and fitness; Step 2.6: Determine whether the maximum number of iterations is met. If not, repeat steps 2.3 to 2.5 until the maximum number of iterations is met. Step 2.7: Output the optimal static obstacle avoidance path for the AUV.

4. The underwater robot path planning method for avoiding obstacles in complex underwater environments according to claim 3 is characterized in that: In step 2.4, the updating of individual optimal position, global optimal position and fitness specifically includes: Step 2.4.1: Introduce the dynamic weight coefficient γ into the discoverer formula to obtain the discoverer dynamic weight coefficient; Step 2.4.2: Introduce the finder ratio coefficient η, and obtain the improved finder position based on the finder dynamic weight coefficient and the finder ratio coefficient η; Step 2.4.3: obtaining the positions of known static obstacles in the grid map during the exploration process based on the improved finder position; Step 2.4.4: Remove the way of approaching the origin during the convergence process of the participant position formula, and only retain the information of the participant approaching the optimal fitness function position, obtain the improved participant position, and obtain the position of the unknown static obstacle in the grid map based on the improved participant position; Step 2.4.5: Update the position of the AUV in the grid map in real time, and update the individual optimal position, the global optimal position and the fitness based on the position of the AUV in the grid map, the positions of the known static obstacles in the grid map and the positions of the unknown static obstacles in the grid map; The expression of the dynamic weight coefficient γ is: The expression of the finder's proportional coefficient η is: The improved calculation formula for the finder's position is: The improved joiner position formula is:

5. The underwater robot path planning method for avoiding obstacles in complex underwater environments according to claim 1, characterized in that: The steps to complete the AUV's dynamic obstacle avoidance in step 3 include: Step 3.1: Initialize the grid map and IDWA algorithm parameters; Step 3.2: Receive dynamic obstacle information detected by the forward-looking sonar and update the initialized grid map; Step 3.3: Sample the velocity of the AUV and generate the trajectory space; Step 3.4: Introduce the AUV's own performance-related threshold D auv Calculate the AUV dynamic weight and update the AUV's motion trajectory; Step 3.5: Use the improved DWA algorithm evaluation function to evaluate the motion trajectory of the updated AUV; Step 3.6: Select the best trajectory among the trajectories of the objects that have not collided with the obstacles according to the evaluation results, and execute the best path; AUV's own performance related threshold D auv The calculation formula is: In formula (6), ξ is a fixed coefficient value used to adjust the relationship between the AUV scale and its speed, D auv The maximum linear velocity u of the AUV max Proportional to linear acceleration Inversely proportional.

6. The underwater robot path planning method for avoiding obstacles in complex underwater environments according to claim 5 is characterized in that: The steps of AUV dynamic weight in step 3.4 include: dynamic speed weight α v , Dynamic heading weight coefficient b v and dynamic distance weight coefficient c v .

7. The underwater robot path planning method for avoiding obstacles in complex underwater environments according to claim 5, characterized in that: The steps for calculating the AUV dynamic weight in step 3.4 include: Step 3.4.1: Based on the AUV's own performance-related threshold D auv and the distance D between the dynamic obstacle and the AUV min Calculate the dynamic speed weight α v ; Step 3.4.2: By comparing the AUV's own performance related threshold D auv and the distance D between the dynamic obstacle and the AUV min Design dynamic heading weight coefficient b v ; Step 3.4.3: Distance D from the AUV through the dynamic obstacle min Design dynamic distance weight coefficient c v ; Dynamic speed weight α v The expression is: Dynamic heading weight coefficient b v The expression is: Evaluation index b closely related to heading angle σ The calculation formula is: ρ = -1·(θ-90) (10); In formulas (9) and (10), b v is the heading dynamic weight coefficient, which is related to the AUV heading information; b σ is an evaluation index closely related to the heading angle. ρ is related to the heading angle of the AUV and is used to distinguish the AUV orientation. θ is the heading angle in the AUV's own coordinate system. max and b min is the heading coefficient of AUV under extreme conditions; Dynamic distance weight coefficient c v The calculation formula is:

8. The underwater robot path planning method for avoiding obstacles in complex underwater environments according to claim 5, characterized in that: The expression of the improved DWA algorithm evaluation function in step 3.5 is: G(v,ω)=a v ·velocity(v,ω)+b v ·heading(v,ω)+c v ·dist(v,ω) (12); In formula (12), ω is the angular velocity of the robot, velocity(ν,ω) is the velocity evaluation function, heading(ν,ω) is the heading evaluation function, and dist(ν,ω) is the distance evaluation function.

Images