Wearable Robot Force Control via Spring-Damper Model
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wearable muscular power assist robots face challenges with inaccurate EMG sensors and sensitivity to object weight variations, making it difficult for users to lift objects with constant force and increasing the burden on the user's body.
Innovation Solution
A method and system using a force sensor to derive a final force for the robot, applying a spring-damper virtual force model to convert this force into torques for the robot's joints, and controlling the robot to maintain a constant force during lifting and transfer operations, utilizing a Jacobian transpose for precise torque calculation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If EMG sensors are used to detect user motion intention, then the robot can be controlled by muscle contraction signals, but the sensing precision deteriorates when the sensor is not closely attached to the user's body
Solution Approach 1:
The patent extracts the force sensing function from the EMG sensor system and implements it through a separate force sensor attached to the robot end effector. This separates the motion detection function (still using EMG) from the force measurement function (using force sensor), allowing each to operate optimally without the limitations of the other.
Solution Approach 2:
The force sensor acts as an intermediary between the user's muscle contraction and the robot's lifting action. Instead of directly measuring muscle electrical signals, the system measures the actual force applied through the force sensor, providing more reliable feedback for controlling the lifting operation.
2Adaptability or versatility
If the robot reacts sensitively to weight variations, then it can adapt to different object weights, but the user experiences varying lifting forces making it difficult to maintain constant force
Solution Approach 1:
The patent implements a feedback control system where the force sensor continuously measures the actual lifting force, and the controller adjusts the robot's actuator output to maintain the desired constant force. The feedback loop compares the measured force with the target force and compensates for weight variations automatically, allowing the user to lift objects of various weights with constant effort.
Solution Approach 2:
The system dynamically changes the control parameters (actuator force output) based on the measured object weight and desired lifting force. By adjusting the robot's assistance force in real-time according to the load conditions, the system maintains constant user effort while adapting to different object weights.
3Extent of automation
If attachment-type EMG sensors are used, then motion intention can be detected, but the device complexity and cost increase making commercialization difficult
Solution Approach 1:
The patent extracts the force measurement function from the complex EMG sensor system and implements it through a simpler force sensor. This separation allows the use of a less complex sensing system while maintaining or improving the overall control capability.
Solution Approach 2:
The force sensor used in the patent is a simpler, more cost-effective component compared to attachment-type EMG sensors. This substitution reduces the overall system cost and complexity, making the wearable robot more commercially viable while still achieving the desired motion control functionality.
Data Source
AI summary
Disclosed herein is a method and system for controlling the lifting operation of a wearable robot. A final force that must be applied by the robot to an object upon conducting a lifting operation is derived based on a difference between a weight force applied by the object to the robot and an apply force applied by a wearing user to the robot. A target position to which the robot lifts the object is set. A spring-damper virtual force model is applied to an end of the robot and to joints of the robot, the final force is converted into final torques required by the joints of the robot by being incorporated into the virtual force model, and then the joints of the robot are operated based on the final torques. The final force is fixed once the robot has lifted the object to the target position.