Compliant Robot Simulation Using Soft Contact and PD Joint Control

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing machine learning-based robotic control approaches face challenges in generating accurate training data, as real-world physical robots require extensive resources, time, and can suffer wear and tear, while simulated data often fails to accurately reflect real-world environments due to the 'reality gap' between simulated and real scenarios.

Innovation Solution

The use of techniques such as compliant end effector models, soft constraints for contact models, and proportional derivative (PD) control in robotic simulators to simulate compliant robotic control and contact, along with system identification for optimizing parameters, to generate more realistic simulated data that bridges the reality gap.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If training data is generated using real-world physical robots, then the accuracy and realism of training data is improved, but resource consumption, time, and wear on robots increase significantly

Engineering Contradiction:

Improveaccuracy of training dataVSAvoidtime for generating training data

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent creates a digital twin (simulated robot) that copies the physical robot's characteristics, dynamics, and environment. This virtual copy generates training data without consuming physical resources, time, or causing wear on the actual robot, while maintaining sufficient realism for effective machine learning training

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The patent introduces a simulator as an intermediary between the physical robot and the machine learning training process. The simulator acts as a mediator that translates real robot characteristics into virtual training scenarios, eliminating the need for extensive physical robot operation while preserving training data quality

Inventive Principle:

Principle #24Intermediary (Mediator)

2Loss of time

If a robotic simulator is used to generate simulated training data, then resource consumption and time are reduced, but the 'reality gap' between simulated and real scenarios deteriorates model performance

Engineering Contradiction:

Improvetime for generating training dataVSAvoidaccuracy of simulated training data

Core Design Contradiction:

Loss of timeVSMeasurement precision

Solution Approach 1:

The patent systematically adjusts simulator parameters including physics engine settings, sensor noise characteristics, friction coefficients, and mass properties to match real robot behavior. By carefully tuning these parameters, the simulated training data closely approximates real-world scenarios, closing the reality gap while maintaining computational efficiency

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent incorporates feedback mechanisms where simulated sensor data and state information are continuously compared with expected real-world behavior. This feedback loop allows the simulator to adjust its output to better reflect actual robot-environment interactions, improving the fidelity of generated training data

Inventive Principle:

Principle #23Feedback

3Measurement precision

If compliant end effector models and soft constraints are implemented in simulation, then the realism of contact simulation is improved, but computational complexity increases

Engineering Contradiction:

Improverealism of contact simulationVSAvoidcomplexity of simulation model

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies compliant contact models and soft constraints selectively at specific contact points between the end effector and environment, rather than throughout the entire robot system. This localized approach maintains high realism where it matters most (contact interactions) while keeping the overall simulation computationally manageable

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS11458630B2Mitigating reality gap through simulating compliant control and/or compliant contact in robotic simulator

Publication Date: 2022.10.04 GDM HOLDING LLC

AI summary

Mitigating the reality gap through utilization of technique(s) that enable compliant robotic control and/or compliant robotic contact to be simulated effectively by a robotic simulator. The technique(s) can include, for example: (1) utilizing a compliant end effector model in simulated episodes of the robotic simulator; (2) using, during the simulated episodes, a soft constraint for a contact constraint of a simulated contact model of the robotic simulator; and/or (3) using proportional derivative (PD) control in generating joint control forces, for simulated joints of the simulated robot, during the simulated episodes. Implementations additionally or alternatively relate to determining parameter(s), for use in one or more of the techniques that enable effective simulation of compliant robotic control and/or compliant robotic contact.