Adaptive Driving Model for Virtual Test Driving
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Determining optimal vehicle setup parameters and driving strategies for racing is challenging due to variations in driver preferences and environmental conditions, as conventional systems struggle to model subjective driver views and adapt to changes in vehicle configurations, leading to suboptimal performance.
Innovation Solution
A hierarchical reinforcement learning approach is employed to generate a driving model that accounts for vehicle setup, driver preferences, and environmental settings, allowing for the determination of optimal vehicle setup parameters and driving strategies through simulations, which can adapt to changes in race characteristics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional systems are used to model driver control, then implementation is simpler, but the ability to adapt to changes in vehicle configurations and determine optimal grip and balance parameters deteriorates
Solution Approach 1:
The driving model uses dynamic parameter adjustment where the driver behavior model continuously adapts grip and balance parameters based on real-time vehicle configuration changes. The model transitions from static pre-programmed behaviors to dynamic adaptive behaviors that respond to varying vehicle setups, allowing optimal performance across different configurations without requiring complete re-modeling.
Solution Approach 2:
The system changes key parameters such as grip coefficient and balance distribution dynamically based on vehicle configuration. By adjusting these parameters in response to different vehicle setups, the model maintains accurate representation of driver behavior across varying conditions without increasing overall system complexity.
2Measurement precision
If detailed driver characteristics are incorporated into the model, then driving behavior accuracy improves, but the difficulty of determining optimal parameters increases
Solution Approach 1:
The driving model performs self-calibration by automatically determining optimal grip and balance parameters through simulated driving trials. Instead of requiring manual tuning based on detailed driver characteristics, the model learns and adjusts parameters autonomously through reinforcement learning from simulated race outcomes, reducing the difficulty of parameter determination while maintaining high accuracy.
Solution Approach 2:
The system incorporates feedback loops where simulated driving performance is evaluated and used to adjust driver behavior parameters. The model receives feedback from virtual race results and continuously refines grip and balance parameters to match observed driver characteristics, achieving high accuracy without manual intervention.
3Productivity
If extensive real-world testing is conducted to optimize vehicle setup, then performance optimization improves, but time consumption and cost increase
Solution Approach 1:
The system creates a virtual copy of the racing environment and vehicle dynamics through high-fidelity simulation. By testing vehicle setups in this virtual replica rather than physical prototypes, the system achieves the same optimization goals without the time loss and costs associated with extensive real-world testing, while maintaining transferability to actual racing conditions.
Data Source
AI summary
Systems and methods are provided for evaluating different vehicle and race configurations in simulation through deriving a driving model that can accurately account for driver preferences, environment settings and changes in vehicle setup without using independent models. The driving model may be used to simulate race scenarios to determine optimal vehicle setup parameters, optimal goals and optimal functions to be implemented on a vehicle during a race. The systems and methods may include generating goals according to training data; determining functions to achieve the goals based on algorithms; training a driving model based on the training data, the goals and the functions; performing one or more simulations of a set of goals and a set of functions derived from the driving model based on first race characteristics; and generating results comprising an optimal vehicle setup, optimal goals and optimal functions according to the simulations.