Human-computer interaction likelihood evaluation method, device and medium based on deep RNN

Through the human-computer interaction likelihood value evaluation method based on deep RNN, the interaction likelihood value of the robot in complex environments is evaluated, which solves the problem of difficulty for operators to quickly select robots that need to be intervened, and improves task efficiency and decision accuracy.

CN119126988BActive Publication Date: 2025-03-18BEIHANG UNIV
2 Cites 0 Cited by

Patent Information

Application Number
CN202411621110.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-11-14
Publication Date
2025-03-18
Estimated Expiration
2044-11-14

AI Technical Summary

Technical Problem

In complex and dynamic environments, it is difficult for a single operator to quickly screen and select robots that require manual intervention, resulting in inefficiency in tasks. The prior art relies on complex sensor systems and precise robotic and environmental models, and its effects are limited.

Method used

Using a human-computer interaction likelihood value evaluation method based on depth RNN, the robot's environmental depth image is obtained when he autonomously navigates motion, and inputs it to the action value estimation network and the interactive likelihood value estimation network, generate the robot's interactive likelihood estimation value, and select the robot with the largest estimated value for control.

Benefits of technology

It improves the decision accuracy and efficiency of human-computer interaction systems, reduces dependence on complex sensor systems and precise models, and is suitable for multi-robot systems in dynamic and complex environments.

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

Abstract

The present application discloses a method, device and medium for evaluating the likelihood value of human-machine interaction based on deep RNN, and relates to the field of human-machine hybrid intelligent technology. The method comprises obtaining an environmental depth image of a robot during autonomous navigation and movement; inputting the environmental depth image into an action value estimation network module to obtain an action value estimation; inputting the environmental depth image into an interactive likelihood estimation network module to obtain an environmental state hidden feature vector; inputting the action value estimation into an interactive likelihood estimation network module to obtain an action value estimation hidden feature vector; inputting the environmental state hidden feature vector and the action value estimation hidden feature vector into a human-machine interaction likelihood estimation module to obtain an interactive likelihood estimation value of the robot, and selecting the robot with the largest estimated value according to the interactive likelihood estimation value of each robot. The present application can actively select a multi-robot system in a dynamic and complex environment.
Need to check novelty before this filing date? Find Prior Art