Robotic Assembly With Adaptive Compliance for Variable Part Poses
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Industrial robotic assembly operations face challenges with variable part positions, requiring accurate execution without precise measurement accuracy, leading to costly and laborious task-specific programming.
Innovation Solution
The implementation of an adaptive assembly strategy that learns from human demonstrations and self-experimentation, utilizing force/torque sensors to correct pose inaccuracies through non-linear compliant control laws, allowing robots to adapt to changes in start and goal poses with minimal programming.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If industrial vision cameras are used to determine part positions, then the robot can operate with variable part positions, but the measurement accuracy is insufficient for precise assembly operations
Solution Approach 1:
The patent introduces force/torque sensors as an intermediary measurement device between the robot and the part. Instead of relying solely on vision cameras for position determination, the force sensors provide tactile feedback that compensates for the insufficient accuracy of visual measurement, enabling precise assembly operations despite variable part positions
Solution Approach 2:
The patent implements a feedback mechanism where force/torque sensors continuously monitor contact forces during assembly operations. This feedback loop allows the robot to adjust its motion in real-time based on actual physical interactions, correcting for initial position inaccuracies detected by vision systems and achieving precise assembly despite measurement limitations
2Manufacturing precision
If dedicated computer programs are developed to modify robot paths for variable part positions, then accurate assembly can be achieved, but the programming cost and complexity increase significantly
Solution Approach 1:
The patent enables the robot system to self-adjust and self-correct during assembly operations through real-time force feedback. Instead of requiring pre-programmed paths for each specific part position variation, the robot autonomously adapts its trajectory based on tactile sensor data, eliminating the need for complex task-specific programming while maintaining assembly accuracy
Solution Approach 2:
The patent transitions from static, pre-programmed robot paths to dynamic, real-time trajectory adjustment. The robot continuously modifies its motion path during execution based on force feedback from sensors, allowing adaptive response to variable part positions without requiring complex offline programming for each scenario
3Ease of operation
If strict repeatability of position and orientation is assumed, then open-loop programs can be executed easily, but the system cannot accommodate variable part positions
Solution Approach 1:
The patent closes the control loop by incorporating force/torque sensors that provide real-time feedback during assembly operations. This feedback mechanism allows the system to maintain simple program execution while adapting to variable part positions, as the sensors automatically compensate for position variations without requiring complex programming modifications
Solution Approach 2:
The patent dynamically changes motion parameters during execution based on force feedback. Instead of fixing all position and orientation parameters in advance, the system allows certain parameters to be adjusted in real-time based on tactile sensor readings, enabling both simple programming and adaptability to variable part positions
Data Source
AI summary
A robot for performing an assembly operation is provided. The robot comprises a processor configured to determine a control law for controlling a plurality of motors of the robot to move a robotic arm according to an original trajectory, execute a self-exploration program to produce training data indicative of a space of the original trajectory, and learn, using the training data, a non-linear compliant control law including a non-linear mapping that maps measurements of a force sensor of the robot to a direction of corrections to the original trajectory defining the control law. The processor transforms the original trajectory according to a new goal pose to produce a transformed trajectory, update the control law according to the transformed trajectory to produce the updated control law, and command the plurality of motors to control the robotic arm according to the updated control law corrected with the compliance control law.