Mobile Body Predictive Control Under Stochastic Noise
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing mobile body control techniques struggle with ensuring accuracy when there is a difference between actual and indicated values, and complex formulae make them impractical for applications like transport robots.
Innovation Solution
A mobile body control apparatus that includes a robot control unit and a control input calculation unit. The control input calculation unit calculates a second control input value for the next control cycle by simulating predictive control input values, taking into account stochastic noise, to minimize route costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If simulation is performed a number of times for each periodically repeated control cycle to find optimal control input sequence, then control adaptability to changing obstacles is improved, but control accuracy cannot be guaranteed when there is a difference between actual indicated value and logical indicated value
Solution Approach 1:
The patent implements feedback by using the actual output values from simulations to update and refine control input values. The control input calculation unit calculates new control inputs based on the relationship between indicated values and actual values obtained from simulation, creating a closed-loop system that continuously improves control accuracy while maintaining adaptability to changing obstacles.
2Reliability
If (risk-sensitive) evaluation function with complex formulae is used to analytically determine control input, then control accuracy is improved, but the formulae are too complex to solve and thus cannot be applied to practical cases such as transport robots
Solution Approach 1:
The patent uses simulation to create a virtual copy of the control system where complex calculations can be performed without requiring complex analytical formulae. By copying the system behavior in a simulated environment and using the results to determine control inputs, the patent achieves accurate control while avoiding the computational complexity of analytical solutions.
Solution Approach 2:
The patent replaces complex analytical mathematical processing with simulation-based iterative calculation. Instead of solving complex risk-sensitive evaluation formulae analytically, the system uses numerical simulation to obtain control inputs, substituting a computationally simpler approach that is more suitable for practical robotic applications.
3Productivity
If weighted average value of control input values that minimize final cost is used, then control optimization is improved, but control accuracy is compromised when actual speed differs from indicated speed
Solution Approach 1:
The patent performs preliminary simulation calculations to establish the relationship between indicated values and actual values before determining the final control input. By pre-calculating the simulation results and using them to adjust the control input values, the system maintains both optimization and accuracy, accounting for discrepancies between indicated and actual values in advance.
Data Source
AI summary
A conventional mobile body control apparatus includes: a robot control unit configured to control a mobile body based on a first control input value; and a control input calculation unit configured to calculate a second control input value that will be the first control input value of the next control cycle, in which the control input calculation unit calculates a route cost through simulation of at least one predictive control input value predicted to be provided to the robot control unit after the first control input value, the simulation being performed by taking into account a stochastic noise, to thereby calculate the second control input value that results in a smaller route cost.