Robot Arm Payload Estimation Using Multi-Orientation Gravity Sensing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users face difficulties in accurately setting payload information for robot arms, leading to improper control and safety issues due to manual entry errors or omissions, as existing methods require precise alignment of force sensors with gravity, which is challenging.
Innovation Solution
A method involving a robot tool flange arranged in multiple orientations relative to gravity, using a force-torque sensor to measure forces and torques, allowing for the calculation of payload mass and pose based on these measurements, thereby automating the payload information acquisition process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If users manually enter payload information into the kinematic model, then the process is simple and quick, but the accuracy and reliability of payload information is poor due to user errors or omissions
Solution Approach 1:
The system performs self-calibration by automatically determining payload information through multiple measurements at different orientations. The robot arm autonomously executes the calibration sequence, moves through predetermined orientations, and processes the measurement data without requiring user intervention, thereby eliminating manual entry errors while maintaining simplicity.
Solution Approach 2:
The calibration method changes the orientation parameter of the robot arm to multiple predetermined orientations during measurement. By measuring forces and torques at different orientations and processing these varying parameters, the system accurately determines payload mass and center of gravity position without requiring manual input.
2Measurement precision
If a force sensor is used to measure payload forces, then measurement capability is provided, but the difficulty of aligning the sensor axis with gravity precisely increases
Solution Approach 1:
Instead of requiring static alignment of the force sensor with gravity, the method dynamically measures forces at multiple different orientations of the robot arm. The sensor remains fixed while the robot arm moves through predetermined orientations, transforming the alignment problem into a dynamic measurement sequence that eliminates the need for precise initial sensor alignment.
Solution Approach 2:
The method transitions from a single-orientation measurement (requiring precise alignment in one dimension) to multi-orientation measurements in three-dimensional space. By measuring forces at multiple orientations and processing the data collectively, the system determines payload parameters without requiring the sensor axis to be precisely aligned with gravity in any single orientation.
3Reliability
If payload information is not accurately set, then the robot arm may drift and safety functions fail, but automated determination methods increase system complexity
Solution Approach 1:
The system performs preliminary calibration by determining payload information before actual robot operations. The automated calibration sequence, including moving through predetermined orientations and processing measurements, is executed in advance to establish accurate payload parameters, ensuring reliable control during subsequent operations without requiring complex real-time adjustments.
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 approach enables precise and easy determination of payload information, improving the accuracy of robot arm control and ensuring proper safety functions by automating the process, even with complex payload forms and less expensive force-torque sensors.
Implementation Method 1
obtaining the force and the torque provided to the robot tool flange by gravity acting on the payload
Data Source
AI summary
Methods and robot, where payload information of a payload attached to a robot tool flange of a robot arm are obtained by arranging the robot tool flange in a plurality of different orientations in relation to gravity; obtaining the force and the torque provided to the robot tool flange by gravity acting on the payload using a force torque sensor arranged at the robot tool flange; obtaining the mass of the payload based on the obtained forces obtained at at least two of the different orientations. The dependent claims describe possible embodiments of the robot and methods according to the present invention.