Universal Balancing Controller for Bipedal Robot Stabilization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current bipedal robot controllers are inadequate for providing robust lateral stabilization in dynamic and unstable environments, as they often require knowledge of the environment and are limited to specific conditions, failing to stabilize robots on surfaces like seesaws and bongo boards.
Innovation Solution
A universal balancing controller is developed as an output feedback controller that uses global robot data, such as pelvis position and link angles, to stabilize bipedal robots in various environments without requiring environmental measurements, enabling stabilization on dynamic surfaces like seesaws and bongo boards as well as static surfaces.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional controllers are used for bipedal robots, then control is effective in stable environments with known properties, but the controller fails to provide robust stabilization in dynamic and unstable environments
Solution Approach 1:
The controller is designed as a universal balancing controller that can stabilize bipedal robots across multiple environment types (static floors, curved surfaces, dynamic unstable surfaces like seesaws and bongo boards) without requiring environment-specific tuning. The controller uses a unified control law that adapts to different environmental conditions through feedback from global robot state measurements, eliminating the need for separate controllers for different environments.
Solution Approach 2:
The controller does not require external environment models or prior knowledge of environmental properties to function. Instead, it autonomously adapts to the current environment by using feedback from global robot state measurements (position, velocity, orientation) and automatically adjusting control inputs to maintain stability, making the system self-sufficient across varying conditions.
2Reliability
If environment-specific controllers are designed for each dynamic environment, then stabilization performance improves for that specific environment, but the overall system complexity increases
Solution Approach 1:
A single universal controller design replaces multiple environment-specific controllers. The controller achieves this by using a unified control law that works across static and dynamic environments through feedback from global robot state measurements, significantly reducing the overall system complexity while maintaining stabilization performance.
Solution Approach 2:
The controller merges the functionality of multiple environment-specific controllers into a single unified control system. By combining the stabilization capabilities for static floors, curved surfaces, and dynamic unstable surfaces into one controller that uses global state feedback, the system achieves simplified architecture without sacrificing performance in any specific environment.
3Measurement precision
If controllers require environmental measurements and models, then control accuracy improves for known environments, but the controller becomes inapplicable to unknown or changing environments
Solution Approach 1:
The controller eliminates the need for external environment models or measurements by using only global robot state measurements (position, velocity, orientation) as feedback. The controller autonomously adapts to any environment by inferring environmental properties from robot state changes and automatically adjusting control inputs, making it versatile across known and unknown environments.
Solution Approach 2:
The controller uses continuous feedback from global robot state measurements to adapt to environmental conditions in real-time. By monitoring changes in robot position, velocity, and orientation, the controller infers environmental properties and adjusts control inputs dynamically, achieving both precision and versatility without requiring explicit environment models.
Data Source
AI summary
A robot, such as a bipedal robot, that includes three or more rigid links such as two legs and a pelvis. The robot includes joints pivotally connecting pairs of the rigid links and an actuator associated with each of the joints. The robot includes a universal balancing controller with an output feedback control module providing control signals to selectively drive the actuators to balance the robot on a support element which may be configured to provide a dynamic, unstable environment or to provide a static, stable environment. During use, the control signals are generated in response to processing of global robot data from sensors associated with the rigid links or the joints. The control signals are generated by the output feedback control module without any need for measurements of the support element or without any measurement of a dynamic environment.