Parameter Estimation Method for Industrial Robots via Fast and Robust Trajectory Design

Parameter Estimation Method for Industrial Robots via Fast and Robust Trajectory Design

This technology improves parameter identification efficiency by collecting robot position and torque data, removing noise using zero-phase low-pass filters and the RLOESS algorithm, and generating optimized excitation trajectories that reduce computational complexity through the use of Hadamard's inequality.

Conventional methods for designing excitation trajectories for robot dynamic parameter estimation have faced challenges with high optimization computational complexity and long processing times as the number of parameters increases.

This technology introduces optimized signal processing steps (zero-phase low-pass filtering and RLOESS smoothing) for position, velocity, acceleration, and torque data, and implements an excitation trajectory generation algorithm with high computational efficiency by applying Hadamard's inequality during the determinant optimization process. Applicable to industrial robots and automation systems, it enhances the accuracy of dynamic parameter estimation while reducing complexity and operational time for parameter optimization.

‍

‍

Key Features:

• Step (S140) of optimizing the trajectory used for estimating the robot's dynamic parameters by applying a least-squares parameter estimator to the results obtained from dynamic estimation modeling.
• Step (S110) of collecting position or torque data of a robot in a parameter estimation method via trajectory design.
• Signal processing step (S120) of reducing noise in the collected data to improve the accuracy of the data.
• Step (S130) of performing dynamic estimation modeling of the robot.

‍

This invention was developed with support from the Ministry of Science, ICT and Future Planning for the development of renewable energy and intelligent robot convergence technology.