Robot Pose Programming Using Surface Normals in Workcells
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Robot programming requires immense manual effort, especially in defining end effector poses and adapting to workcell changes, which is challenging due to the need for precise knowledge of workcell properties.
Innovation Solution
A system that allows developers to generate control instructions for robots using visual indication and intuitive input methods, such as click-and-drag, to define end effector poses and adapt to workcell changes, reducing manual programming and enhancing precision.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional teaching pendant methods are used to program robot movements, then precise control of robot configurations is achieved, but immense manual programming effort and time are required
Solution Approach 1:
The system creates a digital twin or virtual model of the physical workcell, allowing developers to program and simulate robot movements in a virtual environment before deploying to the actual robot. This copying approach preserves the precision needed for real robot control while enabling faster iteration and testing in the virtual model.
Solution Approach 2:
The patent replaces the mechanical teaching pendant interface with a software-based programming interface that runs on a computer. This substitution eliminates the need for physical hand-guiding of the robot while maintaining precise control through software-defined motion paths and configurations.
2Reliability
If detailed manual programming is performed to account for workcell physical properties, then accurate robot task performance is achieved, but the programming process becomes extremely complex and time-consuming
Solution Approach 1:
The system performs preliminary measurements and characterizations of the workcell physical properties during an initial setup phase. Once these properties are captured and stored in the virtual model, they can be reused for multiple programming tasks without requiring repeated manual measurements, thereby maintaining accuracy while reducing programming complexity.
Solution Approach 2:
The virtual workcell model serves multiple functions: it stores physical properties, simulates robot movements, validates programming logic, and generates control instructions. This multi-functionality consolidates what would otherwise be separate complex processes into a single unified system.
3Manufacturing precision
If traditional methods are used to define end effector poses, then precise object manipulation is achieved, but the process requires multiple trial-and-error entries and is time-demanding
Solution Approach 1:
The system provides visual feedback in the virtual environment showing the robot's simulated movements and the end effector's position relative to the target object. This immediate feedback allows developers to see the effect of their programming decisions in real-time, eliminating the need for multiple trial-and-error cycles with physical robots.
Solution Approach 2:
The virtual model pre-calculates and stores the relationship between robot joint configurations and end effector poses. When a developer specifies a desired end effector position in the virtual environment, the system quickly retrieves the corresponding joint configurations without requiring real-time iterative calculation, thereby maintaining precision while dramatically improving speed.
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating control instructions for operating a robot. One of the methods includes generating an interactive user interface that illustrates an object to be manipulated by a robot by using an end effector; receiving, within the user interface, first user input data indicating a workcell location; computing a surface normal of a surface in the workcell corresponding to the workcell location; presenting, within the user interface, a graphical representation of the surface normal corresponding to the workcell location; receiving, within the user interface, second user input data selecting the workcell location; and generating pose data for the robot using the computed surface normal and the workcell location.