Robot Force Control Parameter Tuning Under Load Constraints
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robot systems face challenges in setting appropriate force control parameters due to sensor noise and variations in workpiece shapes and positions, leading to a higher likelihood of exceeding predetermined load ranges during learning processes.
Innovation Solution
A method is introduced that sets a limit value and an objective function with a penalty mechanism to optimize force control parameters, where the penalty increases with exceedance from an allowable value to an upper limit, ensuring the force control characteristic values remain within specified constraints.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a predetermined setting range for force control parameters is established, then learning can be performed with clear reward criteria, but sensor noise and variations cause the load to frequently exceed the setting range, reducing reliability
Solution Approach 1:
The patent transforms the hard constraint problem into a soft constraint optimization problem by changing the parameter representation from binary compliance/non-compliance to a continuous penalty-based objective function. The objective function incorporates penalty terms that increase with the degree of constraint violation, allowing the system to find optimal parameters that minimize violations rather than strictly enforcing hard boundaries that are easily exceeded by sensor noise.
Solution Approach 2:
The patent performs preliminary optimization of force control parameters by pre-calculating the optimal parameter settings that minimize the objective function before actual robot execution. This preliminary action includes determining the penalty weights and constraint boundaries in advance, so that during real-time operation, the system can directly apply these optimized parameters without complex real-time decision-making, thereby improving reliability while reducing operational complexity.
2Reliability
If an allowable value smaller than the limit value is set with penalty mechanism, then the possibility of exceeding the limit value is reduced, but the parameter optimization becomes more complex
Solution Approach 1:
The patent implements feedback through the penalty mechanism in the objective function, where the penalty term provides continuous feedback about the degree of constraint violation. When the force control characteristic value approaches or exceeds the allowable value, the penalty increases, guiding the optimization process to adjust parameters away from violating regions. This feedback loop enables the system to learn and adapt parameter settings that inherently respect constraints without requiring complex real-time monitoring and adjustment mechanisms.
Data Source
AI summary
A method of the present disclosure includes (a) setting a limit value specifying a constraint condition with respect to a specific force control characteristic value detected in force control and an objective function with respect to a specific evaluation item relating to the work, (b) searching for an optimal value of the force control parameter using the objective function, and (c) determining a setting value of the force control parameter according to a result of the searching. The objective function has a form in which a penalty increasing according to an exceedance of the force control characteristic value from an allowable value smaller than the limit value is added to an actual measurement value of the evaluation item.