Moving state monitoring method and device based on Manhattan distance and computer equipment

Abstract

The invention relates to a movement state monitoring method and device based on Manhattan distance and computer equipment. The method comprises the following steps: determining a Manhattan distance between a starting position and a target point position of an intelligent agent in an artificial potential field; when the total step number of the intelligent agent moving from the initial position according to the preset step length exceeds the step number of the intelligent agent moving according to the Manhattan distance and the intelligent agent does not reach the target point, deadlock state monitoring is conducted on the intelligent agent, and position information of the intelligent agent after each time of movement is recorded; after the number of iterations is preset, obtaining the distance between the current position of the current moment and the position before the T moment after each movement of the intelligent agent; if the distance between the current position and the previous position is smaller than a preset threshold value, it is judged that the agent enters a deadlock state. According to the method, the starting time for triggering and starting deadlock state monitoring is determined based on the Manhattan distance, deadlock monitoring does not need to be carried out in the whole process, the operation occupation is reduced, and the path planning efficiency can be improved.

Description

technical field [0001] The present application relates to the technical field of artificial potential field path planning, and in particular, to a method, device, computer equipment and storage medium for moving state monitoring based on Manhattan distance. Background technique [0002] The artificial potential field method has a strong ability to adapt to the scene in practical planning applications, and can display the topology model of the scene. More attention and application. [0003] However, the path planning of the agent by the artificial potential field method has defects and constraints in practical application scenarios. One of them is the local minimum problem, that is, for the problem of multiple obstacles, when the distribution of obstacles conforms to a specific situation, planning In the process, there will be a local extreme value problem, and the agent cannot obtain the global optimum, and then always moves and wanders in this local area. When path planni...

AI Technical Summary

Problems solved by technology

[0003] However, the path planning of the agent through the artificial potential field method has defects and constraints in practical application scenarios, one of which is the local minimum problem, that is, for the problem of multiple obstacles, when the distribution of obstacles conforms to specific conditions, the planning In the process, there will be a local extremum problem, and the agent cannot get the global optimum, and then always moves and wanders in the local area

When using the artificial potential field method for path planning, the judgment of whether the agent is trapped in the local minimum area usually runs through the entire path planning process, which requires considerable calculation

Therefore, the prior art has the problem of low efficiency

Images