Robot Control Program Validation for Predictive Error Prevention
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robot systems face safety concerns due to potential errors in control programs that can lead to hazardous states for humans and the environment, especially when interacting with humans and dynamic environments.
Innovation Solution
A method for predictive testing of control programs to identify and prevent error states by modifying the program, generating warnings, or stopping execution, ensuring the robot system remains within permitted states, utilizing a central control unit, sensors, user interfaces, and processor units connected over a dynamic data network.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the control program is modified during execution to adapt to dynamic environments, then the adaptability of the robot system is improved, but the risk of introducing errors and hazardous states increases
Solution Approach 1:
The system performs preliminary validation of control program modifications before they are executed. The validation unit checks proposed modifications against safety criteria and permitted state definitions in advance, preventing erroneous modifications from being applied to the running control program, thus resolving the contradiction between adaptability and safety.
Solution Approach 2:
The system continuously monitors the robot's current state and compares it against defined permitted states. When a modification is proposed, feedback from the current state analysis informs the validation process, ensuring that modifications only proceed if they maintain the system within safe operational boundaries, balancing adaptability with reliability.
2Reliability
If predictive testing is performed to detect potential error states, then the safety of the robot system is improved, but the execution time and productivity may be reduced
Solution Approach 1:
The validation unit performs partial validation by focusing only on critical safety aspects and permitted state constraints rather than exhaustive testing of all possible execution paths. This selective approach provides sufficient safety assurance while minimizing the time penalty on productivity.
Solution Approach 2:
The system changes the parameter of validation depth based on operational context. For non-critical operations, lighter validation is applied to maintain productivity, while for critical operations involving human interaction or hazardous states, more thorough predictive testing is performed to ensure safety, thus resolving the contradiction dynamically.
Data Source
AI summary
The invention relates to a method for controlling a robot system as well as a robot system. The robot system includes the following components: a robot ROBO with elements driven by actuators; first sensors S1i for sensing a current robot state; a central control unit ZSE, which executes a current control program SP(t) for controlling the robot system; one or more user interfaces NSp; one or more processor units PEr (205), which execute services MPSr for the central control unit ZSE and/or for one or more of the other components of the robot system; wherein the robot ROBO, the first sensors S1i, the central control unit ZSE, the user interfaces NSp, and the processor units PEr communicate with one another over a data network DN. The central control unit ZSE is configured and executed to predictively test whether an execution of the current control program SP(t) will lead to an error state. If such an error state is predicted during the test, execution of one or more actions takes place.