Robot obstacle avoidance guiding method based on Skinner operating condition reflection principle

Abstract

The invention relates to a robot obstacle avoidance guiding method based on the Skinner operating condition reflection principle. The method comprises the steps that firstly, the action set probabilities of a robot are established and are evenly distributed; one action is selected at random, the corresponding position change is calculated, the negative ideal degree corresponding to a new position is worked out according to the distance between the robot and an obstacle and the distance between the robot and a target point, then the orientation function value is obtained, the action probability distribution is adjusted according to the orientation function value and the operating condition reflection principle, and system entropy is calculated; when the system entropy is close to the minimum, the robot advances by the angle at which the action with the largest probability points; the learning process is repeated until the robot arrives at the destination. According to the method, the robot can well simulate the operating condition reflex behaviors of people and animals, the intelligent level of the robot is improved, the robot has strong self-learning, self-organizing and self-adapting capabilities, the robot can explore the environment by itself and successfully avoid the obstacle to perform guiding without an advisor signal.

Description

technical field [0001] The invention relates to the field of obstacle avoidance navigation of mobile robots, in particular to a robot obstacle avoidance navigation method based on the principle of Skinner operating conditional reflex. technical background [0002] In recent years, obstacle avoidance navigation has become a research hotspot in the field of intelligent robots. The purpose of obstacle avoidance navigation is to enable the robot to pass through various obstacles in the environment map, and reach the end point from the starting point safely and without collision. Traditional obstacle avoidance navigation methods can be divided into global information known, global information unknown and partial unknown according to the degree of information provided by the working environment. In information-unknown navigation, the robot needs to use its own perception system to detect the external environment, so as to obtain the location information such as the orientation an...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the problems in the prior art that the intelligence level of robot obstacle avoidance navigation is not high, the present invention proposes a robot obstacle avoidance navigation method based on the Skinner operating conditional reflex principle, so that the robot can " "learning-by-doing" trial-and-error method interacts with the environment, establishes operational conditioned reflexes, and completes obstacle avoidance and navigation

Images