Robotic End-Effector Constraint Control for Object Manipulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Robotic arms face challenges in efficiently manipulating constrained objects in restricted environments without requiring extensive computational resources.
Innovation Solution
A computer-implemented method for robotic arms that determines and executes a path by assigning impedance values along axes of freedom and constrained axes using a task space model, incorporating a task buffer to store and analyze measured task parameters, and a task observer to ensure engagement with the object.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional path planning methods are used for robotic arms manipulating constrained objects, then the manipulation accuracy can be maintained, but the computational complexity and time required increase significantly
Solution Approach 1:
The patent segments the complex path planning problem into two distinct phases: a planning phase that computes the full path, and a tracking phase that executes with simplified impedance control. This segmentation allows complex computations to be performed only when necessary, while execution uses lightweight real-time control, thereby reducing overall computational complexity during manipulation while maintaining accuracy.
Solution Approach 2:
The patent performs preliminary path planning computations before actual manipulation begins. The complete path is calculated in advance using full dynamics models, then this pre-computed path is used as a reference during execution. This preliminary action eliminates the need for complex real-time path recalculation, reducing computational burden during actual manipulation while preserving accuracy through the pre-planned optimal path.
2Manufacturing precision
If complex computational models are used for real-time path planning, then the path accuracy improves, but the real-time execution capability deteriorates
Solution Approach 1:
The patent dynamically switches between two control modes: using full dynamic models during the offline planning phase to ensure path accuracy, and switching to simplified impedance control during real-time tracking to enable fast execution. This dynamic adaptation of computational complexity to the operational phase allows the system to achieve both high path accuracy and real-time execution speed.
3Productivity
If the robotic arm follows a pre-determined path without adaptation, then the computational overhead is reduced, but the ability to handle object constraints deteriorates
Solution Approach 1:
The patent implements feedback through the impedance control mechanism during path tracking. The end effector's interaction forces with the constrained object are continuously measured and used to adjust the control inputs, allowing the system to adapt to actual object constraints while following the pre-computed path. This feedback-based adaptation maintains constraint handling capability without requiring complex real-time path recalculation, thus preserving computational efficiency.
Data Source
AI summary
A computer-implemented method executed by data processing hardware of a robot causes the data processing hardware to perform operations. The robot includes an articulated arm having an end effector configured to engage with an object. The operations include receiving a measured task parameter set for the end effector. The measured task parameter set representing positions of the end effector while manipulating the object. The operations also include generating a task space model for the object based on the measured task parameter set. The task space model modelling the at least one constrained axis of the object. The operations further include limiting movement of the end effector along the at least one constrained axis of the object based on the task space model.