Reducer Parameter Identification with Flexible Joint and Friction Modeling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing dynamic parameter identification methods for collaborative robots fail to accurately consider joint flexibility and friction, leading to reduced modeling accuracy and parameter identification precision.
Innovation Solution
A self-adaptive identification method for nonlinear dynamic parameters of a reducer is proposed, which models each flexible joint as a concatemer of a rigid reducer and an elastic torsion spring, and uses a Fourier series optimized trajectory, a least square method, and a recursive least square method for offline and online parameter identification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If the joint flexibility is ignored in the modeling process, then the modeling process is simplified, but the modeling accuracy is reduced
Solution Approach 1:
The flexible joint is segmented into a rigid reducer part and an elastic element part. The rigid reducer is modeled using traditional rigid body dynamics, while the elastic element is modeled separately using spring-damper characteristics. This segmentation allows the complex flexible joint to be handled through manageable components, maintaining modeling accuracy while controlling complexity.
Solution Approach 2:
The flexible joint is modeled as a composite system combining rigid mechanical components (reducer) and elastic elements (springs and dampers). This composite modeling approach captures both the rigid transmission characteristics and the flexible deformation characteristics, achieving accurate representation of the actual joint behavior without oversimplification.
2Productivity
If the second difference method is used to calculate joint acceleration from rotation angle, then the acceleration can be obtained, but the system noise is amplified and accuracy decreases
Solution Approach 1:
The system uses measured joint torque as feedback to directly calculate acceleration through the dynamic model, rather than relying solely on numerical differentiation of position data. This feedback-based approach compensates for noise in the position measurements and provides more accurate acceleration values.
Solution Approach 2:
The patent replaces the numerical differentiation method (which amplifies noise) with a physics-based dynamic model approach. By using the measured torque and the known dynamic parameters in the equation of motion, acceleration is calculated through physical laws rather than mathematical differentiation, thereby avoiding noise amplification.
3Measurement precision
If a filter is introduced to reduce noise, then the noise is reduced, but input delay occurs and system response speed decreases
Solution Approach 1:
The patent replaces signal filtering methods (which cause delay) with a torque-based acceleration calculation method. By using the measured joint torque and dynamic model to directly compute acceleration, the system avoids the need for filtering position signals, thereby maintaining fast response speed while achieving accurate acceleration measurement.
4Productivity
If the current of the motor is used to estimate joint torque, then the torque can be obtained, but noise in the estimated torque value occurs due to friction influence
Solution Approach 1:
The patent replaces the current-based torque estimation method (which is affected by friction and generates noise) with a torque sensor measurement approach. By directly measuring the joint torque using a torque sensor, the system eliminates the need for friction compensation and current-torque conversion, thereby achieving accurate and noise-free torque data.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This method enables accurate offline identification and online updating of dynamic parameters, improving dynamic modeling accuracy and control effectiveness of collaborative robots, especially in complex working conditions with variable loads and human-robot interaction.
Implementation Method 1
models each flexible joint as a concatemer of a rigid reducer and an elastic torsion spring
Implementation Method 2
uses a Fourier series optimized trajectory, a least square method, and a recursive least square method for offline and online parameter identification
Data Source
AI summary
Provided is a self-adaptive identification method for nonlinear dynamic parameters of a reducer, which belongs to the design field of a reducer. The method includes: modeling a harmonic reducer corresponding to a flexible joint as a concatemer of a rigid reducer and an elastic torsion spring, and carrying out dynamic theoretical modeling and parameter variable independence processing on the concatemer to form a dynamic equation for parameter identification; giving an optimized motion trajectory to each joint of a robot and controlling the robot to act accordingly, acquiring relevant data needed for parameter identification based on a built-in torque sensor and double encoders inside the joint; using an offline identification algorithm to accurately identify a plurality of dynamic parameters of a collaborative robot considering joint flexibility and friction, and obtaining a minimum parameter set.