Tendon-Driven Robot Control With AI Compensation for Precision
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robotic systems with integrated actuators face challenges such as high cost, bulkiness, energy inefficiency, and limited precision due to the mass and inertia of actuators, and tendon-driven mechanisms suffer from imprecision due to variability in tendons, which affects the control of dexterous end effectors.
Innovation Solution
A robotic system is designed with actuators positioned distant from joints along the kinematic chain, using tendon-driven joints and a neural network feedback control loop to compensate for imprecision, allowing for precise control and reduced mass distribution, thereby improving dexterity, safety, and energy efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If actuators are integrated within the mechanical structure near joints to be moved, then the robot can achieve precise control and high power, but the mass and inertia of actuators work against low inertia and fast response, making the robot bulky and energy inefficient
Solution Approach 1:
The patent extracts actuators from the kinematic chain and positions them on the structural frame upstream from joints. This separation removes the heavy actuator mass from the moving links, reducing the moment of inertia and enabling faster, more energy-efficient operation while maintaining actuator power through tendon transmission.
2Weight of moving object
If tendon-driven mechanisms are used to separate actuators from robot links, then link size, mass, and inertia are decreased supporting more dexterity, but variability of tendons introduces imprecision in actuation due to changes in length and elasticity
Solution Approach 1:
The patent implements a neural network feedback control loop that continuously monitors robot state and compensates for tendon variability. The neural network learns and adapts to tendon characteristics, real-time adjustments to counteract imprecision from tendon elasticity and length changes, maintaining high actuation precision despite using lightweight tendon-driven mechanisms.
3Productivity
If actuators are positioned upstream from joints using tendon-driven joints, then dexterity and energy efficiency are improved, but imprecision in tendon translation of actuator motion to robot joints increases with tendon length
Solution Approach 1:
The neural network feedback control loop compensates for position control imprecision by continuously monitoring actual robot state and adjusting actuator commands. The system learns tendon characteristics and predicts compensation values, maintaining high positioning accuracy despite long tendon lengths that would otherwise amplify transmission errors.
Solution Approach 2:
The patent dynamically adjusts control parameters through neural network learning, adapting to tendon variability in real-time. By changing control gains and compensation values based on learned tendon characteristics, the system maintains precision across varying operating conditions and tendon states.
Data Source
AI summary
A robot is provided having a kinematic chain comprising a plurality of joints and links, including a root joint connected to a robot pedestal, and at least one end effector. A plurality of actuators are fixedly mounted on the robot pedestal. A plurality of tendons is connected to a corresponding plurality of actuation points on the kinematic chain and to actuators in the plurality of actuators, arranged to translate actuator position and force to actuation points for tendon-driven joints on the kinematic chain with losses in precision due to variability of tendons in the plurality of tendons. A controller operates the kinematic chain to perform a task. The controller is configured to generate actuator command data in dependence on the actuator states and image data in a manner that compensates for the losses in precision in the tendon-driven mechanisms.