Unmanned ship dynamic obstacle avoidance path planning method based on sailing angle calculation and LOS

Abstract

According to the unmanned ship dynamic obstacle avoidance path planning method based on sailing angle calculation and LOS, the predicted arrival time of the waypoint on the path is obtained while the path is generated. And dynamically adjusting the navigation speed through a dynamic window method, so that the unmanned ship can reach each waypoint according to a plan, thereby realizing collision prediction. When a dynamic obstacle is detected, a new path is quickly planned through an angle pre-compensator according to the relative navigation attitude of the dynamic obstacle and the ship. Compared with other methods, an obstacle avoidance algorithm of the method is derived from ship navigation characteristics, does not depend on machine learning, and is small in calculation amount and wide in adaptability.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of path planning, and particularly relates to an unmanned ship dynamic obstacle avoidance path planning method based on a navigable angle calculation and a LOS. BACKGROUND

[0002] Obstacle avoidance, especially dynamic obstacle avoidance, is one of the difficulties in the practical application of unmanned ships and is also a current research hotspot. Obstacle avoidance is often achieved by planning a reasonable path. Current unmanned ship obstacle avoidance methods can be divided into global path planning-based methods and local path planning methods.

[0003] Global path planning, such as the A* algorithm and the Smac Hybrid-A* algorithm, achieves obstacle avoidance by searching for an obstacle-free path on a cost map. Such an algorithm cannot fully consider the movement of obstacles and can only passively plan a path according to the current position of an obstacle at the current time point, so the dynamic obstacle avoidance capability is poor. If the time dimension is added, the path of an obstacle in the future is predicted, and if the path meets the planned path, the path is re-planned, which can improve the dynamic obstacle avoidance capability. However, the disadvantage is that the meeting time is inaccurate, so the estimated collision point has a large error. Meanwhile, the search of multiple paths has a large amount of calculation, and often requires a server computer to achieve, which is difficult to adapt to the actual scene of a small unmanned ship / boat.

[0004] Local path planning, such as the artificial potential field method and the dynamic window method, has small calculation amount and high real-time performance, but is usually based on the currently detected obstacle position for path planning. The speed obstacle method selects a speed within a speed range that can avoid an obstacle in the future while maintaining the current course and speed as much as possible. However, the algorithm highly depends on the speed information of surrounding obstacles and can only handle local obstacle avoidance, which may fail to solve the problem in the presence of multiple obstacles. In addition, there is an obstacle avoidance algorithm based on deep learning, which learns the optimal obstacle avoidance strategy through continuous interaction (trial and error) with the environment and according to the obtained rewards or punishments. However, the calculation amount is large and the real-time performance is poor.

[0005] Therefore, the application proposes a path planning algorithm based on the combination of navigable angle obstacle avoidance and LOS, which can quickly generate a path. The line-of-sight guidance method (LOS) is a widely used navigation control algorithm in the field of ships, which can control the ship to travel on the planned course, and the calculation is simple and can be realized through an embedded system. The navigable angle obstacle avoidance is based on the field of view principle and selects an unobstructed direction as the navigation angle. The combination of the two can generate a planned path that can avoid obstacles. When there is a dynamic obstacle, the route control based on the dynamic window method can accurately estimate whether it meets the dynamic obstacle, and a dynamic obstacle avoidance algorithm that can pre-compensate the dynamic obstacle near the meeting point is proposed, finally forming an obstacle avoidance strategy for unmanned ships that integrates path planning, collision prediction, and dynamic obstacle avoidance. SUMMARY

[0006] The purpose of this invention is to propose a dynamic obstacle avoidance path planning method for unmanned vessels based on navigable angle calculation and LOS, which can perform dynamic obstacle avoidance with a small computational load.

[0007] To achieve the above objectives, the technical solution of the present invention is: a dynamic obstacle avoidance path planning method for unmanned surface vessels based on navigable angle calculation and LOS, specifically including the following steps:

[0008] S1. Combining the navigable angle obstacle avoidance principle and LOS control process, the ship's trajectory is simulated by a simplified model to obtain the planned path from the starting point to the end point to avoid static obstacles, and at the same time, the estimated arrival time information for each waypoint is obtained.

[0009] S2. Based on the speed information of the dynamic obstacle and the arrival time of the ship's planned route, estimate whether the obstacle will encounter the ship in the near future; if the obstacle encounters the ship, determine the encounter location, and create a virtual obstacle at the encounter location to replan the route.

[0010] Preferably, step S1 is as follows:

[0011] S11. Detect surrounding static obstacles; determine whether the center position of each surrounding static obstacle is less than the obstacle avoidance distance D. If so, obtain the unnavigable angle range caused by the static obstacle; count all unnavigable angle ranges at the current position of the ship.

[0012] S12. Obtain the reference heading angle using the LOS heading control algorithm; if the heading determined by the LOS heading control algorithm is an unnavigable angle, then start from the reference heading determined by the LOS heading control algorithm and search for navigable areas to both sides to find the first navigable angle, which will be used as the final reference heading angle. ;

[0013] S13. Set the ship's average speed based on the ship's navigation mission and navigation environment; based on the ship's average speed and final reference heading angle... Predict the ship's position after time Δt;

[0014] S14. Using the predicted position of the ship as the current position, return to steps S11-S13 to obtain the new heading angle and the predicted position of the ship, until the destination position is reached.

[0015] If the ship's position changes beyond a preset value during the prediction process, the ship's position and corresponding arrival time are recorded to obtain the ship's planned route and the estimated time to reach the waypoints.

[0016] Preferably, step S2 is as follows:

[0017] S21. Control the ship to sail according to the planned route and estimated arrival time using the Dynamic Window Method (DWB), and detect dynamic obstacles;

[0018] S22. Based on the speed and heading information of the detected dynamic obstacles, calculate the possible positions of the dynamic obstacles within a future preset time range, predict whether the ship will encounter the dynamic obstacles within the future preset time range, and obtain the locations where all dynamic obstacles may encounter the ship within the preset time range; for each encounter event determined, add a dynamic virtual obstacle at the encounter location and save the obstacle speed information.

[0019] S23. Based on the added dynamic virtual obstacles, obtain the unnavigable angle range caused by the dynamic virtual obstacles, and compensate for the unnavigable angle range based on the heading angle of the dynamic virtual obstacles relative to the ship.

[0020] S24. Based on the compensated non-navigable angle range, according to the path planning method in step S1, obtain the subsequent planned path and the estimated time to reach the waypoint.

[0021] Preferably, the unnavigable angle range caused by static obstacles or dynamic virtual obstacles is calculated as follows:

[0022]

[0023] in, The unnavigable angle range caused by static or dynamic virtual obstacles. Here, r represents the center coordinates of the static or dynamic virtual obstacle, r is the radius of the obstacle avoidance range, H is the width of the unmanned surface vessel (USV), and d is the distance between the obstacle and the USV. The coordinates are the current position of the unmanned vessel.

[0024] Preferably, the reference heading angle is obtained through the LOS heading control algorithm, and the specific calculation is as follows:

[0025]

[0026] in, These are the current actual coordinates of the unmanned vessel. These are the coordinates of the starting point of the flight path. The coordinates of the destination of the route. For the desired heading angle, y e Ds represents the distance from the flight path, and is the preset forward look-ahead distance. The reference heading angle determined by the LOS heading control algorithm.

[0027] Preferably, the step of basing the ship's average speed and the final reference heading angle on... The predicted position of the ship after time Δt is calculated as follows:

[0028]

[0029] Among them, v c The average speed of the ship. This is the current predicted position of the ship. and The initial values ​​are x and y. These are the current actual coordinates of the unmanned vessel. The coordinates of the ship's position after the predicted time Δt. The ship's current course; the ship's actual course. Based on the final reference heading angle Update:

[0030]

[0031]

[0032] in, The updated current course of the ship. Let T be the ship's angular velocity and T be the preset control period.

[0033] Preferably, the calculation of the possible locations of dynamic obstacles within a preset time range in the future is specifically as follows:

[0034]

[0035] in, The speed at which the obstacle moves dynamically. For the dynamic obstacle at the current time t n Location coordinates, For dynamic obstacle heading, The coordinates of the dynamic obstacle after time Δt.

[0036] Preferably, the prediction of whether the vessel will encounter a dynamic obstacle within a preset time range in the future is as follows:

[0037] Determine the current time t based on the planned path. n The ship's position coordinates And search for the coordinates of the corresponding path points of the ship after time Δt from the planned path. ;

[0038] Then the distance d' between the dynamic obstacle and the ship at a future time t after time Δt is the Euclidean distance between them:

[0039]

[0040] If the distance d' is less than the set obstacle avoidance distance, it is determined that the dynamic obstacle will meet and collide with the ship at a future time t.

[0041] Preferably, the step of obtaining the unnavigable angle range caused by the added dynamic virtual obstacles, and compensating for the unnavigable angle range based on the heading angle of the dynamic virtual obstacles relative to the ship, is as follows:

[0042] The velocity of dynamic obstacles in the northeast celestial coordinate system Projected onto the ship's heading coordinate system:

[0043]

[0044] in, The ship's current course. The velocity of the dynamic obstacle in the ship's heading coordinate system;

[0045] The angle of the dynamic obstacle relative to the ship's heading is:

[0046]

[0047] The direction of travel of the obstacle relative to the ship is determined by the heading angle of the obstacle relative to the ship, in order to identify the unnavigable angle range. Provide compensation:

[0048] If the dynamic obstacle is moving to the left relative to the vessel, the compensated unnavigable angle range is: If the dynamic obstacle is traveling to the right relative to the vessel, the compensated unnavigable angle range is: First adjustment angle The calculation is as follows:

[0049]

[0050] Where d is the distance between the obstacle and the ship's position, and τ is the custom adjustment time;

[0051] If the dynamic obstacle is moving longitudinally relative to the ship, the compensated unnavigable angle range is: Second adjustment angle The calculation is as follows:

[0052] .

[0053] Preferably, the step of determining the direction of travel of the obstacle relative to the ship based on the heading angle of the dynamic obstacle relative to the ship is as follows:

[0054] If the dynamic obstacle is at an angle relative to the ship's heading satisfy If the dynamic obstacle moves to the left relative to the ship, then... A smaller angle was set to avoid shaking;

[0055] If the dynamic obstacle is at an angle relative to the ship's heading satisfy If the dynamic obstacle moves to the right relative to the ship, then the obstacle will move to the right.

[0056] If the dynamic obstacle is at an angle relative to the ship's heading satisfy or If the obstacle is moving longitudinally (towards or away from) the ship, then the obstacle will move relative to the ship.

[0057] Compared with existing technologies, the present invention has the following advantages: (1) By calculating the planned path using formulas, the complex path search process is avoided, saving resources. (2) When generating the path, the predicted arrival time information of the waypoints is included, which allows for a quick determination of whether dynamic obstacles will encounter the ship in the future. (3) By dynamically adjusting the actual sailing speed using the DWB algorithm, the actual route and the predicted route are not only consistent in route but also synchronized in time, ensuring the reliability of the predicted sailing position. Ultimately, the entire sailing process effectively avoids obstacles in a dynamic prediction-execution process. Attached Figure Description

[0058] Figure 1 This is a schematic diagram of the obstacle avoidance principle based on the navigable angle according to an embodiment of the present invention;

[0059] Figure 2 This is a schematic diagram of the heading angle adjustment after obstacle avoidance processing in an embodiment of the present invention;

[0060] Figure 3 This is a schematic diagram of the relative motion between the dynamic obstacle and the ship in an embodiment of the present invention;

[0061] Figure 4 This is a classification diagram of the dynamic obstacle encountering the ship in an embodiment of the present invention;

[0062] Figure 5 This is an overall block diagram of the obstacle avoidance strategy in an embodiment of the present invention;

[0063] Figure 6 This is a demonstration of the obstacle avoidance effect in an embodiment of the present invention. Figure 1 ;

[0064] Figure 7 This is a demonstration of the obstacle avoidance effect in an embodiment of the present invention. Figure 2 ;

[0065] Figure 8 This is a demonstration of the obstacle avoidance effect in an embodiment of the present invention. Figure 3 . Detailed Implementation

[0066] The following is in conjunction with the appendix Figures 1-8 The technical solution of the present invention will be described in detail below.

[0067] This invention proposes a dynamic obstacle avoidance path planning method for unmanned surface vessels based on navigable angle calculation and LOS (Landing Horizon). The complete obstacle avoidance navigation process is as follows: Figure 5 As shown, the specific steps include:

[0068] S1. Combining the navigable angle obstacle avoidance principle and LOS control process, the ship's trajectory is simulated by a simplified model to obtain the planned path from the starting point to the end point to avoid static obstacles, and at the same time, the estimated arrival time information for each waypoint is obtained.

[0069] S2. Based on the speed information of the dynamic obstacle and the arrival time of the ship's planned course, estimate whether the obstacle will encounter the ship in the near future; if the obstacle encounters the ship, determine the encounter location, and create a virtual obstacle at the encounter location to replan the path.

[0070] In this embodiment, the obstacle avoidance principle based on the navigable angle is as follows: Figure 1 As shown, step S1 is as follows:

[0071] S11. Detect surrounding static obstacles; determine whether the center position of each surrounding static obstacle is less than the obstacle avoidance distance D. If so, obtain the unnavigable angle range caused by the static obstacle; count all unnavigable angle ranges at the current position of the ship.

[0072] The obstacle avoidance distance D is determined based on the ship's length, speed, and maneuverability.

[0073] S12. Obtain the reference heading angle using the LOS heading control algorithm; if the heading determined by the LOS heading control algorithm is an unnavigable angle, then start from the reference heading determined by the LOS heading control algorithm and search for navigable areas to both sides to find the first navigable angle, which will be used as the final reference heading angle. ,like Figure 2 As shown;

[0074] S13. Set the ship's average speed based on the ship's navigation mission and navigation environment; based on the ship's average speed and final reference heading angle... Predict the ship's position after time Δt;

[0075] S14. Using the predicted position of the ship as the current position, return to steps S11-S13 to obtain the new heading angle and the predicted position of the ship. Repeat this process until the destination position is reached.

[0076] During the prediction of the ship's position, if the ship's position changes by more than a preset value (1 meter), the ship's position and the corresponding arrival time are recorded to obtain the ship's planned path and the estimated time to reach the waypoint. Since the position of obstacles is taken into account during the planning process, sailing along this path can avoid static obstacles.

[0077] In this embodiment, step S2 is as follows:

[0078] S21. Control the ship to sail according to the planned route and estimated arrival time using the Dynamic Window Method (DWB), and detect dynamic obstacles;

[0079] The specific implementation process of the dynamic window method is as follows: Assume the current position is P. n Find the closest path point from the planned path, and simultaneously search forward from that path point for a time dt (e.g., 5 seconds) to find the target point, denoted as position P. nt The purpose of the dynamic window method is to control the ship from point P within time dt. n Arrival at P nt Assume the current linear velocity of the ship is v. c The direction is θ c angular velocity is ω c Define the acceleration and angular acceleration ranges for the ship; generate various combinations of velocity and angular velocity using the DWB algorithm, and search for the closest P after time dt. nt The combination of velocity and angular velocity at a point should not cause the vessel to encounter an obstacle. The DWB algorithm is relatively mature and is not within the scope of protection of this invention, so it will not be described in detail here. Through the DWB algorithm, the vessel can be effectively controlled to reach a set path point at a set time, thus making it easy to predict whether it will encounter a dynamic obstacle.

[0080] S22. Based on the speed and heading information of the detected dynamic obstacles, calculate the possible positions of the dynamic obstacles within a preset time range (e.g., 20 seconds), predict whether the ship will encounter the dynamic obstacles within the preset time range, and obtain all possible locations where the ship may encounter the dynamic obstacles within the preset time range; for each encounter event determined, add a dynamic virtual obstacle at the encounter location and save the obstacle's speed information; if a collision is detected within the preset time range, replan the path; by predicting in advance, move away from the direction of travel of the dynamic obstacles to improve avoidance efficiency;

[0081] S23. Based on the added dynamic virtual obstacles, obtain the unnavigable angle range caused by the dynamic virtual obstacles, and compensate for the unnavigable angle range based on the heading angle of the dynamic virtual obstacles relative to the ship.

[0082] Considering that the obstacle is dynamically changing, the replanned path will detour around the obstacle, which will take longer to reach the vicinity of the obstacle. Therefore, it is necessary not only to consider the current position of the obstacle, but also to appropriately extend the position of the obstacle in its direction of travel. To this end, this invention proposes an angle-based pre-compensator to handle the impact of obstacle movement in advance.

[0083] S24. Based on the compensated non-navigable angle range, according to the path planning method in step S1, obtain the subsequent planned path and the estimated time to reach the waypoint.

[0084] In this embodiment, the unnavigable angle range caused by static obstacles or dynamic virtual obstacles is calculated as follows:

[0085] (1)

[0086] in, The unnavigable angle range caused by static or dynamic virtual obstacles. Here, r represents the center coordinates of the static or dynamic virtual obstacle, r is the radius of the obstacle avoidance range (including the obstacle's dimensions and the avoidance distance), H is the width of the unmanned surface vessel (USV), and d is the distance between the obstacle and the USV. The coordinates are the current position of the unmanned vessel.

[0087] This invention can use different types of sensors to detect obstacles; for lidar, since it feeds back obstacle information from various angles and directions, it can directly obtain the non-navigable angle.

[0088] In this embodiment, the reference heading angle is obtained through the LOS heading control algorithm (North-East coordinate system), and the specific calculation is as follows:

[0089] (2)

[0090] in, These are the current actual coordinates of the unmanned vessel. These are the coordinates of the starting point of the flight path. The coordinates of the destination of the route. For the desired heading angle, y e Ds represents the distance from the flight path, and is the preset forward look-ahead distance. The reference heading angle determined by the LOS heading control algorithm.

[0091] In this embodiment, the method based on the ship's average sailing speed and the final reference heading angle... The predicted position of the ship after time Δt is calculated as follows:

[0092] (3)

[0093] Among them, v c The average speed of the ship. This is the current predicted position of the ship. and The initial value is x and y, These are the current actual coordinates of the unmanned vessel. The coordinates of the ship's position after the predicted time Δt. The ship's current course; the ship's actual course. Based on the final reference heading angle Update:

[0094]

[0095]

[0096] in, The updated current course of the ship. Let be the ship's angular velocity, and It should be at the ship's maximum angular velocity ω max Within the range, T is the preset control cycle.

[0097] In this embodiment, a uniform velocity model is used. The calculation of the possible positions of dynamic obstacles within a preset time range in the future is as follows:

[0098] (4)

[0099] in, The speed at which the obstacle moves dynamically. For the dynamic obstacle at the current time t n Location coordinates, For dynamic obstacle heading, The coordinates of the dynamic obstacle after time Δt.

[0100] In this embodiment, the prediction of whether the ship will encounter a dynamic obstacle within a preset time range in the future is as follows:

[0101] Determine the current time t based on the planned path. n The ship's position coordinates And search for the coordinates of the corresponding path points of the ship after time Δt from the planned path. ;

[0102] Then the distance d' between the dynamic obstacle and the ship at a future time t after time Δt is the Euclidean distance between them:

[0103] (5)

[0104] If distance If the distance is less than the set obstacle avoidance threshold, it is determined that the dynamic obstacle will meet and collide with the ship at a future time t.

[0105] The relative positions of moving obstacles to the ship are as follows: Figure 3 As shown, there are multiple possibilities; when avoiding obstacles, on the one hand, we should try to go around the obstacle from behind to improve obstacle avoidance efficiency; on the other hand, we should consider the changes in the unnavigable angle area caused by the change in the distance between the obstacle and the ship; the present invention will compensate for the angle from these two aspects.

[0106] In this embodiment, the step of obtaining the unnavigable angle range caused by the added dynamic virtual obstacles, and compensating for the unnavigable angle range based on the heading angle of the dynamic virtual obstacles relative to the ship, is as follows:

[0107] The velocity of dynamic obstacles in the northeast celestial coordinate system Projected onto the ship's heading coordinate system (the ship's heading is the positive x-axis):

[0108] (5)

[0109] in, The ship's current course. The velocity of the dynamic obstacle in the ship's heading coordinate system;

[0110] The angle of the dynamic obstacle relative to the ship's heading is:

[0111] (6)

[0112] The direction of travel of the obstacle relative to the ship is determined by the heading angle of the obstacle relative to the ship, in order to identify the unnavigable angle range. Provide compensation:

[0113] If the dynamic obstacle is moving to the left relative to the vessel, the compensated unnavigable angle range is: If the dynamic obstacle is traveling to the right relative to the vessel, the compensated unnavigable angle range is: First adjustment angle The calculation is as follows:

[0114] (7)

[0115] Where d is the distance between the obstacle and the ship's position, and τ is the custom adjustment time (e.g., 5 seconds).

[0116] If the dynamic obstacle is moving longitudinally relative to the ship, the compensated unnavigable angle range is: Second adjustment angle The calculation is as follows:

[0117] (8).

[0118] like Figure 4 As shown, areas ① and ② represent obstacles moving to the right relative to the ship, while areas ④ and ⑤ represent obstacles moving to the left relative to the ship. Areas ③ and ⑥ represent approaching ships that are approximately on the same course as the ship. When an approaching ship is in areas ①, ⑤, or ⑥, it indicates that the approaching ship has a tendency to leave. Conversely, when an approaching ship is in areas ②, ③, or ④, it indicates that the approaching ship has a tendency to approach.

[0119] To achieve obstacle avoidance more effectively, this invention pre-compensates for the navigable angle range, thereby enabling adaptive adjustments to different encounter scenarios and improving the accuracy of obstacle avoidance.

[0120] In this embodiment, determining the direction of travel of the obstacle relative to the ship based on the heading angle of the dynamic obstacle relative to the ship is as follows:

[0121] If the dynamic obstacle is at an angle relative to the ship's heading satisfy If the dynamic obstacle moves to the left relative to the ship, then... A smaller angle was set to avoid shaking;

[0122] If the dynamic obstacle is at an angle relative to the ship's heading satisfy If the dynamic obstacle moves to the right relative to the ship, then the obstacle will move to the right.

[0123] If the dynamic obstacle is at an angle relative to the ship's heading satisfy or If the obstacle is moving longitudinally (towards or away from) the ship, then the obstacle will move relative to the ship.

[0124] Following the steps outlined above, control the vessel to ultimately reach the destination of the route. The final navigation outcome is as follows: Figures 6-8 As shown. In Figure 6 Initially, based on the static target distribution, an initial route was planned. It can be seen that using the obstacle avoidance strategy of this invention, the planned path correctly avoided the obstacles. At t=22s, the ship detected a potential collision with moving target 1, therefore the route was replanned, bypassing target 1 from behind. Figure 7 At 47s and 71s, it can be seen that the ship detected moving target 2 and moving target 3, and took evasive action in both cases. Figure 8 At 92s, the ship returned to its initial path and finally completed the entire voyage at 140s.

[0125] The above are preferred embodiments of the present invention. Any changes made to the technical solution of the present invention that do not exceed the scope of the technical solution of the present invention shall fall within the protection scope of the present invention.

Claims

1. A method for dynamic obstacle avoidance path planning of an unmanned ship based on a navigable angle calculation and a LOS, characterized in that, Specifically comprising the following steps: S1, combining the sailable angle obstacle avoidance principle with the LOS control process, obtaining the planning path for avoiding static obstacles from the starting point to the end point through a simplified model simulation of the ship's track, and obtaining the expected arrival time information corresponding to each waypoint; S2, estimating whether the dynamic obstacle will meet the ship in a future period of time according to the speed information of the dynamic obstacle and the arrival time of the planned route of the ship; if the dynamic obstacle meets the ship, determining the meeting point and virtually adding an obstacle at the meeting point to re-plan the path.

2. The method of claim 1, wherein the method is based on a navigable angle calculation and LOS dynamic obstacle avoidance path planning for an unmanned ship. Step S1 is specifically as follows: S11, detecting the surrounding static obstacles; judging whether the center position of each static obstacle is less than the obstacle avoidance distance D from the ship, if yes, obtaining the non-sailable angle interval caused by the static obstacle; and counting all the non-sailable angle intervals of the current position of the ship; S12, obtaining a reference heading angle by the LOS heading control algorithm; if the heading angle determined by the LOS heading control algorithm belongs to an un-navigable angle, starting from the reference heading angle determined by the LOS heading control algorithm, searching for a first navigable angle on both sides to find a final reference heading angle ; S13, according to the ship navigation task and the navigation environment, set the average navigation speed of the ship; according to the average navigation speed of the ship and the final reference heading angle , predict the position of the ship after time Δt S14, returning to execute steps S11-S13 with the predicted ship position as the current position to obtain a new heading angle and a predicted ship position until the destination position is reached; In the process of predicting the ship position, if the ship position changes more than a preset value, the position of the ship and the corresponding arrival time are recorded to obtain the planning path of the ship and the expected time of arrival at the passing point.

3. The method of claim 2, wherein the method is based on a navigable angle calculation and LOS dynamic obstacle avoidance path planning for an unmanned ship. Step S2 is specifically as follows: S21, controlling the ship to sail according to the planning path and the expected arrival time by the dynamic window method DWB, and detecting the dynamic obstacle; S22, calculating the possible position of the dynamic obstacle in a future preset time range according to the speed and heading information of the detected dynamic obstacle, predicting whether the ship will meet the dynamic obstacle in the future preset time range, and obtaining the meeting point of all the dynamic obstacles that may meet the ship in the preset time range; each time a meeting event is determined, a dynamic virtual obstacle is added at the meeting point and the speed information of the obstacle is saved; S23, obtaining the non-sailable angle interval caused by the dynamic virtual obstacle according to the added dynamic virtual obstacle, and compensating the non-sailable angle interval based on the heading angle of the dynamic virtual obstacle relative to the ship; S24, obtaining the subsequent planning path and the expected time of arrival at the passing point according to the path planning method of step S1 based on the compensated non-sailable angle interval.

4. The dynamic obstacle avoidance path planning method based on the sailable angle calculation and the LOS of the unmanned ship according to claim 2 or 3, characterized in that, The non-sailable angle interval caused by the static obstacle or the dynamic virtual obstacle is specifically calculated as follows: wherein, an un-navigable angle interval caused by a static obstacle or a dynamic virtual obstacle, a center position coordinate of the static obstacle or the dynamic virtual obstacle, r is a range radius of obstacle avoidance, H is a width of the unmanned ship, and d is a distance between the obstacle and the unmanned ship, a current position coordinate of the unmanned ship.

5. The dynamic obstacle avoidance path planning method based on the sailable angle calculation and the LOS of the unmanned ship according to claim 2 or 3, characterized in that, The reference heading angle obtained by the LOS heading control algorithm is specifically calculated as follows: wherein, is the current actual position coordinate of the unmanned ship, is the starting point coordinate of the route, is the end point coordinate of the route, is the desired heading angle, y e is the deviation distance from the route, Ds is the preset forward distance, is the reference heading angle determined by the LOS heading control algorithm.

6. The dynamic obstacle avoidance path planning method based on the sailable angle calculation and the LOS of the unmanned ship according to claim 2 or 3, characterized in that, The final reference course angle is calculated according to the average sailing speed of the ship and the final reference course angle The position of the ship after the prediction time Δt is calculated as follows: wherein v c is the average sailing speed of the ship, is the current predicted position of the ship, and the initial values of x and y, is the current actual position coordinate of the unmanned ship, is the predicted position coordinate of the ship after the predicted time Δt, is the current heading of the ship, the current heading of the ship is updated according to the final reference heading angle is updated according to the final reference heading angle wherein, the updated current heading of the ship, is the angular velocity of the ship, and T is a preset control period.

7. The method of claim 3, wherein the method is based on a navigable angle calculation and LOS dynamic obstacle avoidance path planning for unmanned ships. The possible position of the dynamic obstacle in a future preset time range is specifically calculated as follows: wherein, is the dynamic obstacle movement speed, is the position coordinate of the dynamic obstacle at the current time t n is the dynamic obstacle heading, is the position coordinate of the dynamic obstacle after the time Δt.​ 8. The method of claim 7, wherein the method is based on a navigable angle calculation and LOS dynamic obstacle avoidance path planning for an unmanned ship. The prediction of whether the ship will meet the dynamic obstacle in the future preset time range is specifically calculated as follows: According to the planning path, determine the current time t n the ship position coordinates , and search for the corresponding path point coordinates of the ship after time Δt from the planning path ; The distance d' between the dynamic obstacle and the ship at a future time t after time Δt is the Euclidean distance between them: If the distance d' is less than the set obstacle avoidance distance, it is determined that the dynamic obstacle will meet the ship at the future time t and a collision will occur.

9. The method of claim 3, wherein the method is based on a navigable angle calculation and LOS dynamic obstacle avoidance path planning for unmanned ships. The non-sailable angle interval caused by the dynamic virtual obstacle is obtained according to the added dynamic virtual obstacle, and the non-sailable angle interval is compensated based on the heading angle of the dynamic virtual obstacle relative to the ship, which is specifically calculated as follows: The velocity of dynamic obstacles in the northeast celestial coordinate system Projected onto the ship's heading coordinate system: wherein, is the current heading of the ship, is the dynamic obstacle velocity in the ship heading coordinate system; The heading angle of the dynamic obstacle relative to the ship is: The running direction of the dynamic obstacle relative to the ship is determined according to the heading angle of the dynamic obstacle relative to the ship, so as to compensate for the non-navigable angle interval ​ If the dynamic obstacle moves left relative to the ship, the compensated non-navigable angle interval is ; if the dynamic obstacle moves right relative to the ship, the compensated non-navigable angle interval is ; and the first adjustment angle is calculated as follows: Wherein, d is the distance between the obstacle and the ship position, τ is the self-defined adjustment time; If the dynamic obstacle moves longitudinally relative to the ship, the non-navigable angle interval after compensation is , and the second adjustment angle is calculated as follows: 。 10. The method of claim 9, wherein the method is based on a navigable angle calculation and LOS dynamic obstacle avoidance path planning for an unmanned ship. The running direction of the obstacle relative to the ship is determined according to the heading angle of the dynamic obstacle relative to the ship, and the specific process is as follows: If the dynamic obstacle is relative to the heading angle of the ship satisfies then the dynamic obstacle is relative to the left running of the ship, wherein A small angle is set to avoid jitter If the dynamic obstacle is moving in the same direction as the ship satisfies then the dynamic obstacle is moving to the right of the ship; If the dynamic obstacle is relative to the heading angle of the ship satisfies or then the dynamic obstacle is relative to the longitudinal direction of the ship.

Images