Haptic Robot Task Learning for Force-Controlled Teaching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional robot technologies struggle to perform complex and varied tasks, particularly those requiring force information, due to limitations in existing teaching methods such as the teaching pendant method, kinesthetic teaching, teleoperation, and hard coding.
Innovation Solution
A robot system that uses a haptic manipulating device to transfer force information, allowing an expert to demonstrate tasks and enabling the robot to learn through haptic feedback, sensor data, and computer-aided motion learning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional teaching methods (teaching pendant, kinesthetic teaching, teleoperation, hard coding) are used, then robot can perform simple and repetitive tasks rapidly and precisely, but robot cannot perform complex tasks requiring force information and flexible adaptation
Solution Approach 1:
A haptic manipulating device is introduced as an intermediary between the expert operator and the robot. This device transmits force information from the expert's manipulation to the robot, enabling the robot to learn complex tasks requiring force control without complex programming. The haptic device serves as a mediator that captures subtle force interactions during expert demonstration and transfers them to the robot for reproduction.
Solution Approach 2:
The system copies the expert's manipulation behavior through haptic feedback and sensor data. By recording the force information, motion trajectories, and operational context during expert demonstration, the system creates a replicable model that the robot can execute. This copying approach allows complex force-controlled tasks to be transferred from expert to robot without traditional programming complexity.
2Loss of information
If teleoperation method is used, then robot can be controlled remotely, but force information cannot be contained in task because only information on person's posture is transferred
Solution Approach 1:
The haptic manipulating device provides real-time haptic feedback to the expert operator during task demonstration. This feedback loop allows the expert to feel the forces being applied by the robot, enabling intuitive adjustment of manipulation strategies. The feedback mechanism ensures that force information is captured and transmitted accurately, solving the information loss problem of conventional teleoperation while maintaining ease of operation through natural haptic interaction.
3Loss of information
If kinesthetic teaching method is used, then robot can be guided through motion, but robot has difficulty in discerning whether force transmitted from person is for teaching or is required for task
Solution Approach 1:
The system segments the force information into distinct components: teaching forces (applied by the expert to guide the robot) and task forces (required for performing the actual task). Through sensor data analysis and haptic feedback patterns, the system distinguishes between these force types, allowing the robot to learn the appropriate task forces while filtering out teaching-specific forces. This segmentation enables accurate acquisition of force-controlled task information.
Data Source
AI summary
The present invention relates to methods for learning a robot task and robots systems using the same. A robot system may include a robot configured to perform a task, and detect force information related to the task, a haptic controller configured to be manipulatable for teaching the robot, the haptic controller configured to output a haptic feedback based on the force information while teaching of the task to the robot is performed, a sensor configured to sense first information related to a task environment of the robot and second information related to a driving state of the robot, while the teaching is performed by the haptic controller for outputting the haptic feedback, and a computer configured to learn a motion of the robot related to the task, by using the first information and the second information, such that the robot autonomously performs the task.