Unlock instant, AI-driven research and patent intelligence for your innovation.

Real-time multi-mode accompanying robot intention understanding method and system

A companion robot and multi-modal technology, applied in the direction of neural learning methods, instruments, computer components, etc., can solve the problems of rare old people's intention understanding model, and the lack of a good solution to the efficient extraction of environmental semantic information and intention understanding, to achieve The effect of fast running speed, small training data and strong flexibility

Pending Publication Date: 2022-02-25
UNIV OF JINAN
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Based on the relevant research mentioned above, it is not difficult to find that the current research on robot intent understanding and prediction has not solved the problem of efficiently extracting environmental semantic information for intent understanding, and there are few robot researches at this stage that are aimed at the elderly. Intent understanding model for this specific group

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Real-time multi-mode accompanying robot intention understanding method and system
  • Real-time multi-mode accompanying robot intention understanding method and system
  • Real-time multi-mode accompanying robot intention understanding method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0051] Embodiment 1 of the present invention proposes a real-time multi-modal escort robot intent understanding method, part of which is contextual intent reasoning based on knowledge-based semi-naive Bayesian, which mainly includes gesture and body gesture labels obtained in real time by the neural network and detection Intention reasoning results are obtained by fusing the existence features of items with special semantics; the other part is distance-based intention reasoning, including the analysis of the position-distance relationship between characters and special items, and the probability of the user’s intention to interact with related items around is obtained from The intent probabilities derived from distance inference optimize the knowledge-based probabilities so that the probabilities from the two parts are fused to obtain the final intent. Such as figure 1 It is an overall block diagram of the implementation of a real-time multimodal escort robot intent understand...

Embodiment 2

[0085] Based on the method for understanding the intent of a real-time multi-modal escort robot proposed in Embodiment 1 of the present invention, Embodiment 2 of the present invention also proposes a system for understanding the intent of a real-time multi-modal escort robot. Such as image 3 It is a connection schematic diagram of a real-time multi-modal escort robot intent understanding system according to Embodiment 2 of the present invention. This includes a first calculation module, a second calculation module and an intent fusion module;

[0086] The first calculation module is used to identify the user’s local information and the target object when the distance between the user and the robot meets the requirements of safety detection; measure the sum of the distances from all target objects to the user, and calculate the probability of intentional interaction between the target object and the user ;

[0087] The second calculation module is used to obtain the interde...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention provides a real-time multi-mode accompanying robot intention understanding method and system. The method comprises the steps that local information of a user and a target object are recognized; calculating the sum of the distances from all the target objects to the user, and calculating the probability of intention interaction between the target objects and the user; obtaining interdependence information among the intention features; the interdependence information comprises action information and gesture information of the user and target object information stored in a knowledge base in advance, inputting the interdependence information into a semi-naive Bayesian classifier to obtain probability results of all intentions, and performing probability normalization to obtain probabilities of all intentions related to a target object; and performing probability fusion on the probability of intention interaction of the user and the probability of each intention related to the target object to obtain the final probability of each intention. Based on the method, the invention further provides a real-time multi-mode accompanying robot intention understanding system, the real-time multi-mode accompanying robot intention understanding method has good robustness, and the efficiency and accuracy of understanding the user intention by the robot are greatly improved.

Description

technical field [0001] The invention belongs to the technical field of multimodal intent understanding, and in particular relates to a real-time multimodal escort robot intent understanding method and system. Background technique [0002] Aging is currently a problem faced by many countries around the world. Due to the busy work of children, it is difficult to provide parents with the care they need all the time. At the same time, through the actual investigation of nursing homes and research on escort robots, this paper finds that robot escorts are increasingly recognized by the elderly, and elderly escort robots provide many services. However, since the recognition rate and intention understanding rate of the robot escort system for the elderly need to be improved, the interaction burden of the elderly is increased when the elderly use the escort robot. [0003] In recent years, predecessors have done a lot of in-depth research on multi-modal information fusion technology...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06V40/10G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06N3/045G06F18/24155G06F18/256
Inventor 冯志全豆少松田京兰
Owner UNIV OF JINAN