Vehicle Controller Selection Under Input Uncertainty
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle control systems face inaccuracies and uncertainties in selecting controllers due to unreliable data, leading to inappropriate or unsafe controller selection.
Innovation Solution
A system that acquires input values representing uncertainties in variables, simulates vehicle performance for multiple controllers, determines a total cost based on simulations, and selects the most optimal controller based on these costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If controller selection is based on predicted performance using input data, then controller selection accuracy is improved, but reliability deteriorates due to uncertainties in the input data
Solution Approach 1:
The system performs preliminary simulations before actual controller selection by testing multiple controllers with multiple sets of input values that represent uncertainty ranges. This preliminary action allows the system to identify which controllers are robust to input uncertainties before making the final selection, thereby improving reliability while maintaining selection accuracy.
Solution Approach 2:
The system changes the parameters used in controller evaluation by not using single fixed input values but rather multiple input values representing uncertainty ranges. This parameter change approach allows the system to evaluate controller performance under various possible conditions, leading to more reliable selections that account for data uncertainties.
2Reliability
If multiple simulations are performed for each controller to account for uncertainties, then reliability of controller selection is improved, but computational complexity increases
Solution Approach 1:
The system performs simulations with multiple input values for each controller, but not exhaustively for all possible values. Instead, it uses a representative set of input values that capture the essential uncertainty ranges. This partial action approach provides sufficient reliability improvement without the excessive computational burden of exhaustive simulation.
Solution Approach 2:
The simulations are performed in advance before the actual controller selection decision is needed. This preliminary computation allows the system to pre-evaluate controller robustness, and the results can be stored and used for quick selection when needed, thereby reducing the computational burden at the time of actual decision-making.
Data Source
Figure 1
Figure 2~3
AI summary
A computer system for selecting a vehicle controller for use in control of a vehicle, the computer system comprising processing circuitry configured to acquire a plurality of input values for each of one or more input variables for simulating performance of the vehicle, wherein the plurality of input values represent an uncertainty in the respective input variable, simulate performance of the vehicle for each of a plurality of vehicle controllers, wherein a respective simulation is performed for each vehicle controller for each of a plurality of combinations of input values, determine a total cost for each vehicle controller based on the plurality of simulations performed for the respective vehicle controller, and select a vehicle controller from the plurality of vehicle controllers based on the determined costs.