Vehicle Setup Modeling With Driver Feedback and Weather Inputs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional learned models in motorsports fail to account for variable disturbance factors like temperature, humidity, and wind, and do not consider driver feedback when proposing vehicle setups, leading to discrepancies between predicted and actual performance.
Innovation Solution
An information processing device that incorporates feedback information from drivers and environmental data to update learned models, proposing optimal vehicle setups considering handling characteristics and constraints, using a processor to input course, setup, and driver feedback to a learned model for output-based recommendations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional learned models are used to predict optimal vehicle setups, then prediction speed is fast, but prediction accuracy deteriorates due to ignoring disturbance factors like temperature, humidity, and wind
Solution Approach 1:
The learned model is segmented into multiple independent models, each specialized for predicting specific disturbance factors (temperature, humidity, wind speed, wind direction) rather than predicting all factors in a single monolithic model. This segmentation allows each model to focus on specific environmental variables, improving overall prediction accuracy while maintaining computational efficiency through modular architecture.
Solution Approach 2:
The prediction system transitions from a single-dimensional output (optimal setup parameters) to multi-dimensional output by incorporating predictions from multiple disturbance factor models. The system evaluates optimal setups across different environmental conditions simultaneously, adding the dimension of environmental variability to the prediction process, which resolves the contradiction between speed and accuracy.
2Adaptability or versatility
If all variables are handled as variables in learned models, then the model flexibility is high, but the model cannot respect user constraints on specific variables
Solution Approach 1:
The system dynamically adjusts the treatment of variables based on user constraints. When constraints are specified, the system transitions from treating all variables as flexible parameters to fixing constrained variables at their specified values while optimizing only the unconstrained variables. This dynamic adaptability allows the model to switch between flexible exploration and constraint-respecting optimization modes.
Solution Approach 2:
The system changes the parameter treatment mode based on user input. When constraints are provided, constrained parameters are fixed at specific values rather than being optimized as variables. This parameter change approach maintains model flexibility for unconstrained variables while ensuring reliability by respecting user-specified constraints on important parameters.
3Measurement precision
If disturbance factors are considered as constant in learned models, then the model simplicity is maintained, but the difference from actual measured values increases
Solution Approach 1:
The model is segmented into multiple specialized sub-models, each predicting a specific disturbance factor (temperature, humidity, wind speed, wind direction) based on track conditions and vehicle setup. This segmentation allows the system to capture complex environmental variations without requiring a single overly complex model, thus improving alignment with actual measured values while managing complexity through modular design.
Solution Approach 2:
The system performs preliminary predictions of disturbance factors before determining optimal vehicle setups. By first predicting environmental conditions (temperature, humidity, wind) and then using these predictions to optimize setup parameters, the system accounts for disturbance factor variations in advance, improving accuracy without adding excessive complexity to the optimization process itself.
Data Source
AI summary
The information processing device includes a processor, and the processor inputs information on a course on which the vehicle travels, setup information on a setup of the vehicle, and feedback information from a driver driving the vehicle into a learned model, and proposes an optimal setup when the vehicle travels on the course based on an output from the learned model.