Dynamic path planning method for sweeping robot

The environment information is obtained through sensors, distinguish between static and dynamic obstacles, use the motion model to predict the location of dynamic obstacles, define the speed control space, calculate the improved artificial potential field, and combine the reinforcement learning model to optimize the evaluation function, solve the path planning problem of sweeping robots in dynamic environments, and improve obstacle avoidance and cleaning efficiency.

CN120370936APending Publication Date: 2025-07-25DONGGUAN PINJIA INTELLIGENT TECH CO LTD
0 Cites 2 Cited by

Patent Information

Application Number
CN202510463273.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-04-14
Publication Date
2025-07-25

AI Technical Summary

Technical Problem

The existing sweeping robot path planning method has high computational complexity and poor adaptability when dealing with dynamic obstacles, which is difficult to meet real-time requirements, and it is difficult to apply deep reinforcement learning in dynamic environments.

Method used

Acquire environmental information through sensors, distinguish between static and dynamic obstacles, use motion models to predict the location of dynamic obstacles, define the velocity control space, calculate the improved artificial potential field, combine the reinforcement learning model to optimize the evaluation function, and dynamically adjust the path planning.

Benefits of technology

It improves the ability to predict and avoid obstacles for dynamic obstacles, can better deal with dynamic environments, avoid local optimal solutions, and improves cleaning efficiency and energy utilization.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention discloses a sweeping robot dynamic path planning method comprising the following steps: using a sensor to obtain environment information, using a motion model to predict the future position of a dynamic obstacle, defining the motion space of a robot as a speed control space, calculating an improved artificial potential field comprising a gravitational field and a repulsive force field, and defining a prediction collision window to predict the dynamic path of the robot. And constructing a reinforcement learning model, designing a reward function, and integrating a reinforcement learning model network to the sweeping robot. The method solves a problem that a conventional artificial potential field method is liable to fall into a local minimum value, improves the prediction and obstacle avoidance capabilities of dynamic obstacles, optimizes the evaluation function of the DWA algorithm through reinforcement learning, enables the evaluation function to have better adaptability and intelligence, can better process the dynamic environment, can learn the global information of the environment through reinforcement learning, and improves the prediction and obstacle avoidance capabilities of the dynamic obstacles. And the information is integrated into local path planning, so that falling into a local optimal solution is avoided, a path can be rapidly adjusted, it is ensured that a whole cleaning area can be efficiently covered, and the cleaning efficiency and the energy utilization rate are effectively improved.
Need to check novelty before this filing date? Find Prior Art