CN-121979197-A - Unmanned ship autonomous cluster formation obstacle avoidance navigation method

CN121979197ACN 121979197 ACN121979197 ACN 121979197ACN-121979197-A

Abstract

The unmanned ship autonomous cluster formation obstacle avoidance navigation method mainly comprises the steps of initializing parameters, positions, heading and operation plans of each ship in an unmanned ship cluster, sharing navigation information to a cluster network by a pilot ship and a following ship in the navigation process, judging whether an obstacle is met according to sea area environment information detected by a sensing module of the unmanned ship in the navigation process, and performing control behavior conversion by the unmanned ship formation when the obstacle is met, and taking corresponding obstacle avoidance measures. According to the method, through obstacle avoidance path planning of the pilot boat, local collision avoidance based on a dynamic window of the following boat and sequencing according to formation parameters when the whole formation is in obstacle avoidance, unmanned boat clusters can flexibly adopt respective optimal control behaviors to avoid collision on the basis of ensuring safety and accessibility when encountering obstacles, and can be quickly aggregated after obstacle avoidance, cluster navigation is continuously kept, and obstacle avoidance efficiency is effectively improved.

Inventors

QIAO PENGFEI
YANG XIN
FENG DONGLIANG
WANG ZHUO

Assignees

宜昌测试技术研究所

Dates

Publication Date: 20260505
Application Date: 20251222

Claims (9)

1. The unmanned ship autonomous cluster formation obstacle avoidance navigation method is characterized by comprising the following steps of: s1, initializing the formation position, heading and operation plan of each boat in an unmanned boat cluster; s2, sharing navigation information to a cluster network by a pilot boat and a following boat in the navigation process; S3, detecting sea area environment information according to a perception module of the unmanned ship in the navigation process, judging whether the unmanned ship encounters an obstacle, converting control behaviors of the unmanned ship formation when the unmanned ship encounters the obstacle, and taking corresponding obstacle avoidance measures, wherein the piloting ship navigates according to a navigation plan, the navigation plan comprises an initially set navigation plan and an obstacle avoidance plan when the unmanned ship encounters the obstacle, the formation autonomously maintains ordering and safety intervals of the ships according to initially set formation coordinate parameters in the cluster obstacle avoidance process, and the single following ship performs behavior control and obstacle avoidance through a local collision avoidance algorithm.
2. The method according to claim 1, wherein the step S1 comprises: Setting pilot boats of the unmanned boat cluster, boat numbers of following boats, operation plans of the pilot boats and formation position parameters; the method comprises the steps of setting formation coordinates of a pilot boat as (0, 0), taking the right of the pilot boat as a transverse axis X, taking the right of the pilot boat as a positive, taking the front of the pilot boat as a longitudinal axis Y, and taking the front of the pilot boat as a positive; the formation coordinates remain unchanged during sailing, and the following boats are ordered according to the formation coordinates.
3. The method according to claim 1 or 2, wherein the step S2 specifically comprises: Unmanned boats carry wireless communication equipment, and all boats among clusters form a cluster network together; The unmanned ship determines whether the unmanned ship is a pilot ship or not and a ship number needing to share information according to the information sent by the comprehensive display and control console; Each unmanned ship obtains the position longitude and latitude, the course and the speed thereof by measuring through an integrated navigation system; The unmanned ship shares navigation information through a cluster network, wherein the shared navigation information comprises the heading, the speed, the longitude and latitude of the position, the transverse and longitudinal offset and the current control behavior state of the unmanned ship; the pilot boat control behavior comprises sailing along a sailing plan and sailing along an evasion plan, and the following boat control behavior comprises the steps of maintaining formation, avoiding local collision and sailing to a position required by the pilot boat.
4. The method according to claim 3, wherein in the step S3, the pilot boat completes the navigation behavior control and obstacle avoidance through a path planning algorithm and a dynamic window method two-layer collision avoidance algorithm, and the method specifically comprises: The path planning algorithm judges whether collision threat exists in a section of track in real time according to the current planned speed of the pilot boat, and if the collision threat does not exist, the behavior of the pilot boat is 'sailing along a sailing plan'; if the obstacle threat exists, the behavior of the pilot boat is changed into 'sailing along an avoidance plan', and at the moment, a path planning rule rules out an avoidance route which leaves an original route, safely avoids the obstacle and finally returns to the original route; the two layers of collision avoidance algorithms run in parallel and in real time.
5. The method according to claim 4, wherein in the step S3, the pilot boat navigates along the route formed by the navigation plan, and in the navigation process, the navigation speed of the pilot boat is adjusted in real time according to the shared position information and the control behavior state information of the following boat, so that the following boat can form a formation quickly, and the navigation information of the following boat can be shared into the cluster network; and when the pilot boat is positioned at the inflection point of the navigation plan, the pilot boat is decelerated to navigate, and the pilot boat waits to form a formation along with the boat.
6. The method according to claim 4 or 5, wherein during the cluster obstacle avoidance process, the formation autonomously maintains the fore-and-aft sequence and interval of the boats according to the initially set formation coordinates (x, y) of each boat, namely, firstly sorting according to the x-axis size, and then sorting according to the y-axis size under the condition that the x values are the same; after sequencing, when the distance between the rear boat and the front boat is smaller than the safe distance, the distance between the rear boat and the front boat is increased through speed reduction; and recovering the formation after the obstacle avoidance is finished.
7. The method of claim 6, wherein the following boat adopts a dynamic window algorithm in the local collision avoidance to control the sailing behavior and avoid the obstacle during sailing, and the method specifically comprises the following steps: Under normal conditions, the following boat adopts a 'keep formation' action, calculates a guiding target point according to the position of the pilot boat and formation coordinate parameters, and controls navigation to the target point by adopting a dynamic window method; If threat exists at the target point position calculated according to the formation according to the detection of the perception module, the control action is changed into 'local collision avoidance' or 'navigation to the position required by the piloting boat', and then the target point is moved to a safe position to avoid the obstacle.
8. The method according to claim 7, wherein in step S3, the information interaction and navigation control behavior state between the boats under different conditions specifically includes: (1) When the pilot boat and the following boat have no collision threat, the pilot boat behavior state is 'sailing along a sailing plan', the following boat behavior state is 'holding formation', The navigation craft uses the planned speed as the benchmark to match with the following craft to keep the formation and share the navigation information, the following craft calculates the guiding target point according to the position of the navigation craft and the formation coordinate parameter, adopts the dynamic window method to control the navigation to the target point, and outputs the proper speed and heading value by evaluating the predicted course of the speed heading in the set time window to control the unmanned craft to reach the target point and avoid the obstacle; (2) When the piloting boat has no collision avoidance threat, the following boat encounters the collision avoidance threat, the behavior state of the piloting boat is 'sailing along a sailing plan', 'local collision avoidance', The following vessels execute collision avoidance behavior, guide target points are calculated according to the position and formation parameters of the pilot vessels, and the navigation to the target points is controlled by adopting a dynamic window method to avoid threat obstacles; (3) When the pilot boat encounters collision threat and the following boat has no collision threat, the pilot boat behavior state is 'sailing along an avoidance plan', 'keeping formation', The pilot boat sails along an avoidance route by taking the avoidance planning navigational speed as a reference, the shared position point of the following boat to the transverse team is a collision prevention end point, and the shared position point of the following boat to the longitudinal team is the current position point of the boat; The following boat calculates a guiding target point according to the shared position point and the formation coordinate parameter of the pilot boat, and controls navigation to the target point by adopting a dynamic window method; (4) When the pilot boat has collision avoidance threat and the following boat has collision avoidance threat, the behavior state of the pilot boat is 'sailing along an avoidance plan', 'local collision avoidance', Navigation is carried out by the pilot boat along an avoidance route by taking the avoidance planning navigational speed as a reference, a shared position point of the following boat for a transverse team is taken as a collision prevention end point, and a shared position point of the following boat for a longitudinal team is taken as a current position point of the pilot boat; (5) When the pilot boat is in the sailing along the avoidance plan and then finishes avoidance, a tandem boat exists in the following boat, and the last boat in the tandem boat does not reach the collision prevention end point sent by the pilot boat, the behavior state of the pilot boat is the sailing along the avoidance plan, the behavior state of the following boat is the sailing towards the position required by the pilot boat, the pilot boat in the state carries out speed-reducing sailing by taking the planned sailing speed as a reference so as to ensure that the distance between the pilot boat and the following boat is not increased, the shared position point of the pilot boat to other boats is the current position of the pilot boat, the shared position point of the pilot boat to the last boat is the collision prevention end point, and the pilot boat is switched to the behavior state of keeping the formation after the last boat sails to the collision prevention end point.
9. The method according to claim 8, wherein in the step S3, when the pilot sails to the last leg of the initial sailing plan and the distance from the last waypoint is smaller than the set threshold, the pilot completes the plan, and the pilot notifies the following boat of completion of the formation plan, and shares the last waypoint position and heading to the following boat; After the following boats receive formation plan completion information sent by the pilot boat, the following boats calculate guiding target points according to the final position points sent by the pilot boat, when the following boats navigate to the target points, the following boats switch task modes into waiting modes, and when all the following boats reach the respective target points, the completion of formation tasks is indicated.

Description

Unmanned ship autonomous cluster formation obstacle avoidance navigation method Technical Field The invention relates to the technical field of unmanned water surface vessel navigation planning control, in particular to an unmanned water vessel autonomous cluster formation obstacle avoidance navigation method. Background The most widely applied algorithms in the existing unmanned ship formation control method are three, namely a behavior method, a virtual structure method and a pilot following method. The basic principle of the behavior method is that a group of behaviors are designed for each unmanned ship, each behavior has corresponding behavior weight, the control output of the unmanned ship in the final formation is from the weighted average of each behavior, and common behavior designs comprise driving to a target point, preventing collision, maintaining formation, changing formation, aggregating and the like. The basic principle of the virtual structure method is that the whole formation is regarded as a virtual rigid body, unmanned vessels are assumed to be a point fixed on the rigid body and are called virtual structure points, when the formation moves, each unmanned vessel keeps the formation by tracking the corresponding virtual structure points, however, the formation transformation is inflexible and unfavorable for obstacle avoidance when the formation control is carried out by the method. The basic principle of the navigator following method is that one unmanned ship is selected as a navigator in a formation system, the rest unmanned ships are used as followers, the decision of the movement track of the whole formation is made by the navigator, when the unmanned ship serving as the navigator drives to a target point, the following ships can realize formation by adjusting the relative position relation with the navigator, and when the method is used, the organization relation among the unmanned ships in the system is clear, and the formation tasks can be rapidly arranged. In the cooperative formation movement of the unmanned ships, the unmanned ships inevitably encounter obstacles, the unmanned ships need to be subjected to local real-time collision prevention control, and the currently commonly used effective local collision prevention control algorithm mainly comprises a dynamic window method, an artificial potential field method and a speed obstacle method. In order to realize the safe sailing of the multi-unmanned-ship cluster formation in the complicated and changeable seafood environment, a formation control algorithm and a local collision prevention algorithm are required to be combined, and the multi-unmanned-ship autonomous cluster formation sailing is a main effective algorithm at present. However, the formation control algorithm and the local collision avoidance algorithm are directly combined, so that complex and changeable ocean environments and various task modes cannot be dealt with: for example, the virtual structure method and the local collision avoidance algorithm are combined to control the unmanned ship cluster, when the formation requires a long distance between the ships, and when the threat of an obstacle is faced, a certain ship in the cluster can keep the formation at an extremely high navigational speed, and a larger path is required to bypass to finish collision avoidance, so that the execution efficiency of tasks is greatly reduced, energy is consumed, and the cruising ability is reduced; If a pilot following method and a local collision prevention control algorithm are combined to control an unmanned ship cluster, when a pilot faces obstacle threats and needs to be avoided, the following ship needs to change a moving target point for a plurality of times according to the position of the pilot to keep the formation, the navigation track is complex and invalid, the course of the following ship is changed frequently, and the execution efficiency of tasks is reduced; for example, when the number of unmanned ships is large and the unmanned ships face a complex environment, problems of inferior decision, limited formation holding precision, complex management, high parameter adjustment difficulty and the like caused by behavior conflict can occur. Disclosure of Invention Aiming at the problems, the unmanned ship autonomous cluster formation obstacle avoidance navigation method is provided by extracting the advantages of the behavior method and the piloting following formation control method and combining with a dynamic window local collision avoidance algorithm, and the unmanned surface ship cluster formation autonomous obstacle avoidance navigation problem is solved under the unknown marine environment. The method mainly comprises the following steps: s1, initializing parameters, positions, heading and operation plans of each boat in an unmanned boat cluster; s2, sharing navigation information to a cluster network by a pilot boat and a following boat in the navigation p