A method to benchmark the balance resilience of robots

Area of Science:

• Robotics
• Control Systems
• Mechanical Engineering

Background:

• Robots operating in unstructured environments face frequent collisions and external disturbances.
• Developing robust and resilient robotic systems is a critical research area.

Purpose of the Study:

• To present a framework for quantitatively assessing the balancing resilience of self-stabilizing robots under external perturbations.
• To provide novel performance indicators (PIs), experimental protocols, and a dedicated testbed for repeatable measurements.

Main Methods:

• Development of a novel framework including specific Performance Indicators (PIs).
• Establishment of experimental protocols for reliable and repeatable PI measurement.
• Design and implementation of a dedicated testbed for executing perturbation tests.
• Open-sourcing of the testbed design, control structure, software, and documentation for replication.

Main Results:

• A framework for quantitative assessment of robot resilience was successfully developed.
• Experimental validation was performed on a two-wheeled humanoid robot through over 1100 tests.
• The methodology demonstrated high repeatability and efficacy in executing precise perturbations.

Conclusions:

• The proposed framework provides a reliable method for evaluating the balancing resilience of self-stabilizing robots.
• The open-source nature of the framework facilitates wider adoption and replication in the research community.
• This work contributes to the advancement of robust robot design for unpredictable environments.