Robotic Load Estimation Using Precomputed Identification Trajectories

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing robotic manipulators face challenges in accurately and efficiently estimating dynamically changing payloads in real-time due to the complexity and time-consuming nature of nonlinear optimization problems, making fast and flexible load identification infeasible.

Innovation Solution

A supervised learning-based approach is employed to generate optimal identification trajectories for robotic manipulators, using basis functions and offline optimization to predict load parameters efficiently, allowing for real-time load estimation by approximating trajectories with supervised learning methods and rescaling to meet motion constraints.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If nonlinear optimization problems are solved to generate optimal identification trajectories, then measurement precision of load parameters is improved, but loss of time increases due to computation intensity

Engineering Contradiction:
Improveload parameter estimation accuracyVSAvoidtrajectory generation time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent pre-computes and stores optimal identification trajectories for various initial configurations offline before actual load estimation tasks. When online estimation is needed, the system retrieves pre-computed trajectories based on current configuration rather than solving optimization problems in real-time, thus maintaining high measurement precision while dramatically reducing time loss.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent segments the configuration space into discrete regions and pre-computes trajectories for each region separately. This segmentation allows the system to handle complex nonlinear optimization by breaking it into manageable offline computations, enabling fast online retrieval without sacrificing accuracy.

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If optimal identification trajectories are generated through complex optimization, then measurement precision improves, but device complexity increases

Engineering Contradiction:
Improveload parameter estimation accuracyVSAvoidtrajectory generation system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The complex optimization computations are performed in advance during system setup or offline calibration phases. The results are stored in lookup tables or databases, transforming the complex computational problem into a simple retrieval operation during actual use, thereby improving measurement precision without increasing operational device complexity.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

Instead of solving optimization problems repeatedly, the system creates copies of optimal trajectories for different configuration regions and stores them for rapid access. This copying approach replaces complex real-time computation with simple data retrieval, reducing device complexity while maintaining precision.

Inventive Principle:
Principle #26Copying

3Measurement precision

If traditional offline optimization methods are used for trajectory generation, then measurement precision is maintained, but productivity decreases due to slow adaptation to dynamic changes

Engineering Contradiction:
Improvepayload estimation accuracyVSAvoidonline estimation speed
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The system performs preliminary computation of optimal trajectories for all anticipated initial configurations offline. When dynamic payload changes occur during operation, the system immediately retrieves the appropriate pre-computed trajectory based on the current configuration, maintaining measurement precision while achieving fast adaptation that significantly improves productivity compared to real-time optimization.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system dynamically adapts to changing payloads and configurations by selecting from pre-computed trajectories appropriate to the current state. This dynamic selection approach allows the system to respond rapidly to changing conditions without performing complex optimization online, thereby improving both productivity and adaptability while maintaining precision.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP4313512B1Systems and methods for flexible robotic manipulation by fast online load estimation
Publication Date: 2026.02.04 MITSUBISHI ELECTRIC CORP
  • EP4313512B1 patent drawingFigure 1A
  • EP4313512B1 patent drawingFigure 1B
  • EP4313512B1 patent drawingFigure 1C

AI summary

Provided herein is a method for controlling a manipulator, comprising accepting an initial pose of a load and a task for moving the load and retrieving using a mapping function, an identification trajectory corresponding to the initial pose of the load and controlling a plurality of actuators of the manipulator to move the load according to the retrieved identification trajectory and obtaining measured motion data and estimated motion data of the load each corresponding to motion of the load. The method further comprises estimating parameters of the load based on the measured motion data and the estimated motion data, obtaining a model of the manipulator having the load with the estimated parameters, and determining a performance trajectory to move the load according to the task based on the obtained model of the manipulator. The method further comprises controlling the actuators to move the load according to the performance trajectory.