Systems and methods for flexible robotic manipulation by fast online load estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current robotic manipulators face challenges in accurately and efficiently estimating dynamic payloads in real-time, especially in scenarios where payloads and configurations change frequently, due to the complexity of nonlinear optimization problems and the need for fast trajectory generation.
Innovation Solution
The proposed solution involves a supervised learning-based approach for online trajectory generation, where the manipulator's control system uses a load identifier to estimate payload parameters by constructing an optimal identification trajectory based on initial joint configurations, and then processes motion data to solve for load parameters efficiently.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If nonlinear optimization problems are solved for trajectory generation in load estimation, then accurate payload estimation is achieved, but computation time increases and real-time performance deteriorates
Solution Approach 1:
The patent pre-computes and stores optimal trajectories in a database before online operation. When load estimation is needed, the system retrieves pre-computed trajectories based on current joint configurations rather than solving nonlinear optimization problems in real-time. This preliminary action separates the computationally intensive optimization from the time-critical online estimation, resolving the contradiction between accuracy and speed.
Solution Approach 2:
The patent creates a database of pre-computed optimal trajectories that serve as copies of the solutions to nonlinear optimization problems. Instead of repeatedly solving the same optimization problems online, the system copies and retrieves appropriate trajectories from the database based on matched joint configurations. This copying approach maintains estimation accuracy while eliminating real-time computation delays.
2Measurement precision
If optimal trajectories are computed for each load estimation instance, then estimation accuracy is improved, but system complexity and processing load increase
Solution Approach 1:
The system performs trajectory optimization offline and stores results in a database. During online operation, it simply retrieves pre-computed trajectories by matching joint configurations, avoiding the need to solve complex nonlinear optimization problems in real-time. This preliminary computation reduces online processing complexity while maintaining estimation accuracy.
Solution Approach 2:
The patent creates a database of copied optimal trajectories from offline computations. Instead of重新 computing trajectories online, the system copies appropriate trajectories from the database based on current joint configurations. This copying strategy maintains accuracy while dramatically reducing processing complexity during online operation.
3Measurement precision
If traditional trajectory generation methods are used for load estimation, then accurate payload parameters are obtained, but the method is not suitable for online implementation due to time consumption
Solution Approach 1:
The patent pre-computes optimal trajectories offline and stores them in a database before online operation. During online load estimation, the system retrieves pre-computed trajectories based on current joint configurations rather than solving optimization problems in real-time. This preliminary action enables fast online estimation while maintaining payload parameter accuracy.
Solution Approach 2:
The system creates a database of copied optimal trajectories from offline computations. During online operation, it copies appropriate trajectories from the database based on matched joint configurations instead of重新 computing them. This copying approach maintains payload parameter accuracy while enabling fast online estimation suitable for real-time applications.
Data Source
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.