Robot Gaze Sequencing for Multi-User Interaction Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robots struggle to interact effectively with multiple users simultaneously, often focusing on a single user or interacting randomly, leading to inadequate engagement for all users, especially when new users are recognized later and join the interaction group.
Innovation Solution
A robot equipped with a camera, driver, and processor that determines a gaze order based on user interaction scores and locations, rotating to sequentially engage multiple users by identifying target regions within its field of view and adjusting the gaze order to include new users, ensuring all users feel actively interacted with.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If the robot focuses on a single user or interacts randomly, then the interaction control is simple, but the engagement quality for all users deteriorates
Solution Approach 1:
The robot segments the interaction process by dividing users into different groups (first group of users and second group of users) and applying different interaction strategies to each group. The robot determines a gaze order for the first group, interacts with them sequentially, then updates the gaze order to include the second group. This segmentation allows the robot to maintain simple control logic while improving engagement quality for all users.
Solution Approach 2:
The robot dynamically updates the gaze order based on newly recognized users. When a second group of users is recognized, the robot updates the existing gaze order to include these new users, transforming the static interaction sequence into a dynamic one that adapts to changing environmental conditions. This enables the robot to maintain simple operational control while achieving versatile engagement across multiple user groups.
2Adaptability or versatility
If the robot sequentially gazes at multiple users based on a determined gaze order, then the engagement quality for all users improves, but the device complexity increases
Solution Approach 1:
The robot performs preliminary actions by determining a gaze order for the first group of users before actually interacting with them. This pre-planning of the interaction sequence allows the robot to systematically engage multiple users in a structured manner, improving engagement quality while managing control complexity through advance organization of the interaction flow.
Solution Approach 2:
The robot uses feedback from camera-based user recognition to dynamically update the gaze order. When new users are detected, the system receives feedback about their presence and adjusts the interaction sequence accordingly. This feedback mechanism enables the robot to maintain high engagement quality across multiple users while keeping the control system manageable through event-driven updates rather than continuous complex processing.
3Adaptability or versatility
If the robot updates the gaze order to include new users during rotation, then the adaptability to new users improves, but the loss of time in interaction increases
Solution Approach 1:
The robot employs periodic action by updating the gaze order at specific intervals - specifically when new users are recognized during rotation. Rather than continuously adjusting the interaction sequence, the system performs discrete updates at meaningful moments (when new users enter the field of view), thereby improving adaptability to new users while minimizing time loss by avoiding constant re-planning of the interaction sequence.
Data Source
AI summary
A robot and a controlling method thereof are provided. The robot includes: a camera; a driver; at least one memory storing instructions; and at least one processor operatively connected to the at least one memory. The at least one processor may be configured to execute the instructions to: recognize at least one user located around the robot based on an image acquired through the camera; identify a first rotation angle and a first rotation direction of the robot based on a location of at least one target region in which the at least one user is located among a plurality of regions within a field angle of the camera and a first gaze order for the at least one target region; and control the driver to rotate the robot based on the first rotation angle and the first rotation direction.