CN-115981324-B - Unmanned ship autonomous cruising method and device

CN115981324BCN 115981324 BCN115981324 BCN 115981324BCN-115981324-B

Abstract

The application is suitable for the technical field of unmanned ships and provides an unmanned ship autonomous cruising method, which comprises the steps of dividing a preset water area into square matrixes and acquiring environment data in the preset water area, wherein the environment data comprises water area data and obstacle data; setting the attribute information of the square corresponding to the water area data as a water area square, setting the attribute information of the square corresponding to the obstacle data as an obstacle square, setting the attribute information of the water area square adjacent to the obstacle square as an equidistant square, traversing all squares in the preset water area range, generating a square track unit set by using the equidistant square, acquiring a situation matrix taking the position of the unmanned ship as the center at each moment, generating a cruising track according to the situation matrix, and cruising according to the cruising track. According to the scheme, the grid track units are gathered to avoid various obstacles, so that the unmanned aerial vehicle can cruise autonomously.

Inventors

LIU HAO
ZHANG YUNFEI
CHEN DONGLIANG
DENG KE

Assignees

珠海云洲智能科技股份有限公司

Dates

Publication Date: 20260505
Application Date: 20221228

Claims (9)

1. An unmanned aerial vehicle autonomous cruise method, comprising: Dividing a preset water area into square matrixes, and acquiring environment data in the preset water area, wherein the environment data comprises water area data and obstacle data; Setting attribute information of squares corresponding to the water area data as water area squares, setting attribute information of squares corresponding to the obstacle data as obstacle squares, and setting attribute information of the water area squares adjacent to the obstacle squares as equidistant squares; Traversing all squares in the preset water area range, and generating a square track unit set by using the equidistant squares; Acquiring a situation matrix taking the position of the unmanned ship as the center at each moment, wherein the situation matrix comprises the water area square and the obstacle square; Sequentially inquiring the squares in the situation matrix in the square track unit set, and updating the attribute information of the squares in the situation matrix according to the inquiry result, wherein when the squares in the situation matrix are inquired in the square track unit set, the attribute information of the squares is updated to be equidistant squares, otherwise, the attribute information of the squares is kept unchanged; outputting the situation matrix after the inquiry is finished; And generating a cruising track according to the situation matrix, cruising according to the cruising track, executing an obstacle avoidance instruction when the attribute information is equidistant square grids, wherein the cruising track runs along the equidistant square grids, and keeping the distance between the cruising track and an obstacle unchanged.
2. The method of claim 1, wherein the obstacle data comprises static obstacle data and dynamic obstacle data, the obstacle square comprising a static obstacle square and a dynamic obstacle square, the method further comprising: and when the obstacle data corresponding to the square in the situation matrix is dynamic obstacle data, updating the attribute information of the square into the dynamic obstacle square.
3. The method of claim 1, wherein the generating a cruise trajectory from the situational matrix comprises: judging the attribute information of a square corresponding to the position of the unmanned aerial vehicle in the situation matrix, generating a cruising track according to the attribute information of the square, and cruising; Maintaining a forward command to the target site when the attribute information is a water square, or And executing a waiting instruction when the attribute information is a dynamic barrier square.
4. The method of claim 3, wherein executing the obstacle avoidance instruction when the attribute information is equidistant squares comprises: Determining a direction of rotation from a direction to the target site; determining the next travelling square according to the mark; acquiring the center point of the next travelling square, and advancing towards the center point; Judging whether to continue executing the obstacle avoidance instruction according to the attribute information of the next travelling square, Ending the obstacle avoidance instruction when the attribute information is a water area square, or And when the attribute information is not the water area square, continuing to execute the obstacle avoidance instruction.
5. The method of claim 4, wherein determining the next travel square based on the indicia comprises: Judging whether the square corresponding to the position of the unmanned ship is marked or not; When the square is not marked, judging whether the unmanned ship reaches the center point of the square, When the central point is not reached, the method proceeds to the central point of the square or Marking the square when the center point has been reached; When the square is marked, the executing step obtains the center point of the next travelling square and advances towards the center point.
6. The method of claim 5, wherein said obtaining a center point of said next traveling square, proceeding toward said center point, comprises: Acquiring attribute information of squares around the unmanned ship according to the rotation direction; executing a waiting instruction when the information of the water area square is not acquired; When the information of the unlabeled water area square is acquired, advancing to the unlabeled water area square; When all the acquired water area squares are marked, the water area squares with the earliest marking time are advanced.
7. The method of claim 1, wherein the method further comprises: and executing the return and supply instruction after the energy reserve of the unmanned ship falls into a preset range.
8. An unmanned system comprising a grandmaster side and an unmanned device side or only an unmanned device side, the grandmaster side and/or the unmanned device side comprising an intelligent computing device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the method according to any of claims 1-7 when executing the computer program.
9. A computer readable storage medium storing a computer program, which when executed by a processor implements the method according to any one of claims 1-7.

Description

Unmanned ship autonomous cruising method and device Technical Field The application relates to the technical field of unmanned boats, in particular to an unmanned boat autonomous cruising method and device. Background Unmanned vessels are unmanned surface vessels, often equipped with advanced control systems, sensor systems, communication systems, etc., and are primarily used to perform a variety of dangerous tasks and tasks not applicable to the performance of manned vessels. In order to ensure that the unmanned cluster can automatically cruise in the ocean, the unmanned ship must be capable of automatically avoiding various static obstacles and dynamic obstacles encountered in the navigation process, and simultaneously, the unmanned ship should automatically execute energy supply when the energy is insufficient. In the traditional scheme, the method mainly comprises the following steps of taking an obstacle as an avoidance point in path planning based on a path planning algorithm, performing task allocation and path optimization together, processing the obstacle as image data based on a machine learning algorithm or a reinforcement learning algorithm, acquiring characteristic information, autonomously selecting an avoidance strategy, executing an obstacle avoidance action, and improving the obstacle avoidance success rate by repeatedly iterating and accumulating obstacle avoidance experience data, and the method is based on a rule constraint algorithm, wherein a mathematical model of obstacle avoidance is established through simple rules, such as a manual potential field, boids algorithm, a speed obstacle method, a modeling obstacle avoidance algorithm based on the international obstacle avoidance rule at sea, and the like, and the obstacle avoidance action is realized through the mathematical model. However, the method has poor effect of processing the moving obstacle based on the path planning algorithm, long training process and low efficiency based on the machine learning algorithm depending on the live-action obstacle image data, and the rule constraint algorithm based on the rule is simple and difficult to adapt to complex conditions. Disclosure of Invention The embodiment of the application provides an autonomous cruising method of an unmanned ship, which can solve the technical problems of how to realize high-efficiency obstacle avoidance and long-term cruising of the unmanned ship during cruising. In a first aspect, an embodiment of the present application provides an autonomous cruise method of an unmanned ship, including: Dividing a preset water area into square matrixes, and acquiring environment data in the preset water area, wherein the environment data comprises water area data and obstacle data. The attribute information of the square corresponding to the water area data is set as a water area square, the attribute information of the square corresponding to the obstacle data is set as an obstacle square, and the attribute information of the water area square adjacent to the obstacle square is set as an equidistant square. Traversing all the squares in the range of the preset water area, and generating a square track unit set by using equidistant squares. The grid track unit set comprises track routes which can realize obstacle avoidance in the cruising process of the unmanned ship. And acquiring a situation matrix taking the position of the unmanned ship as the center at each moment, wherein the situation matrix comprises water area square grids and barrier square grids. Sequentially inquiring the squares in the situation matrix in the square track unit set, updating the attribute information of the squares in the situation matrix according to the inquiry result, and outputting the situation matrix after the inquiry is finished. And generating a cruising track according to the situation matrix, and cruising according to the cruising track. The cruise track is generated by utilizing the situation matrix on the basis of the generated grid track unit set to implement obstacle avoidance action, so that the vehicle can walk close to the edge of the obstacle to realize the walk, and the vehicle can be well adapted to the limitation of the obstacles such as concave obstacles In one implementation, sequentially querying the squares in the situation matrix in the square track unit set, updating attribute information of the squares in the situation matrix according to the query result, and including: When the squares in the situation matrix can be queried in the square track unit set, updating the attribute information of the squares to be equidistant squares, otherwise, keeping the attribute information of the squares unchanged. In one example, the obstacle data includes static obstacle data and dynamic obstacle data, the obstacle square includes a static obstacle square and a dynamic obstacle square, the method further comprising: And when the obstacle data corresponding to the square in the situa