Drive-By-Wire Control Adaptation for Varying Driver Competence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Advanced driver assistance systems (ADAS) face challenges in accommodating drivers with varying levels of competence and driving styles, as they struggle to seamlessly transition from physical control inputs to drive-by-wire systems, potentially leading to unsafe driving scenarios.
Innovation Solution
A networked system that includes vehicle computing systems, electronic control units, and sensors, which use machine learning and AI to adjust control signals based on driver input patterns, transforming them to match a predetermined safe-driver model, thereby improving the vehicle's control mechanisms to simulate competent driving.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If drive-by-wire systems are implemented to automate vehicle control, then vehicle automation and safety are improved, but the system struggles to accommodate drivers with varying competence levels and driving styles
Solution Approach 1:
The control system dynamically adjusts the transformation function based on real-time analysis of driver input patterns. The system continuously adapts to the driver's competence level and driving style, modifying control signals accordingly rather than using a fixed transformation approach.
Solution Approach 2:
The system changes the parameters of the control transformation by analyzing driver input patterns and adjusting the transformation function to match a predetermined safe-driver model. This allows the same physical control inputs to produce different control signals depending on the driver's competence level.
2Reliability
If the vehicle control system is designed for competent drivers, then driving safety is improved, but drivers with lower competence may find the system unaccommodating and potentially unsafe
Solution Approach 1:
The system introduces an intermediary transformation function between the driver's physical control inputs and the actual vehicle control signals. This intermediary layer analyzes the driver's input patterns and transforms them into control signals that would be appropriate for a competent driver, effectively mediating between the driver's intentions and safe vehicle operation.
Solution Approach 2:
The system creates a virtual model of competent driving behavior and uses this as a template to transform control signals. By copying the input patterns of competent drivers and using them as a reference model, the system can guide less competent drivers toward safer operating patterns while still responding to their actual inputs.
3Reliability
If the system transforms control signals to match a safe-driver model, then driving safety is improved, but the system complexity increases due to machine learning and pattern analysis requirements
Solution Approach 1:
The system performs self-training by automatically analyzing driver input patterns and adjusting its transformation function without requiring external programming or manual configuration. The system learns from the data it collects and improves its performance autonomously, reducing the need for complex external training mechanisms.
Solution Approach 2:
The system implements feedback loops where the transformed control signals and driver inputs are continuously analyzed to improve the transformation function. This feedback mechanism allows the system to learn from actual driving patterns and refine its safety transformations over time, making the complexity manageable through iterative improvement.
Data Source
AI summary
A system can train a vehicle electronically to accommodate a driver. The system can train the vehicle to accommodate the ability, condition, and/or personality of the driver. The system can change the controls of the vehicle, responsive to the inputs from the driver, to match with the patterns of controls resulting from a predetermined model (such as a safe-driver model). Accordingly, the vehicle can appear as it is being driven by a safe driver when it may not be the case. A driver with a lower driving competence may apply physical controls in a pattern that may be slow, unstable, or insufficient. However, the vehicle can be trained to adjust the transformation from the UI signals to the drive-by-wire signals such that the transformed signals appear to be applied by a more competent driver on the road. And, the transformation can improve over time with training via machine learning.