Vehicle Torque Control Using Driver Propensity and External AI Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electric vehicle power control devices face limitations in computational performance, preventing the use of advanced AI algorithms for optimal torque control due to low-end hardware, which affects driver satisfaction and safety, especially with external variables like weather and road conditions.
Innovation Solution
A vehicle driving control apparatus that uses a mobile device and autonomous driving control to enhance computational performance without redesigning existing hardware, by predicting energy requirements and adjusting torque based on driver propensity and situational data using learning algorithms, and sharing data with nearby vehicles for improved control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If computational performance of existing controller is increased, then AI algorithms can be operated for optimal torque control, but safety of legacy software code cannot be guaranteed
Solution Approach 1:
The control system is divided into two independent parts: a legacy control unit that executes verified safety-critical code and a new AI processing unit that handles optimization tasks. This segmentation allows computational enhancement through AI while preserving the safety guarantees of legacy software by maintaining clear functional boundaries between the two systems.
Solution Approach 2:
An intermediary communication interface is introduced between the legacy controller and the new AI processing unit. This intermediary layer enables data exchange and coordinate control actions without requiring modification of the legacy software, thus maintaining its safety certification while allowing AI algorithms to enhance torque control performance.
2Productivity
If hardware calculation performance is improved to digest AI algorithms, then optimal determination result can be achieved, but existing hardware and software must be significantly improved
Solution Approach 1:
The AI processing functionality is extracted from the existing vehicle controller and implemented as a separate processing unit. This extraction allows the use of advanced AI algorithms for optimal torque determination without requiring significant improvement or redesign of the existing hardware platform, as the AI functions are handled by dedicated processing equipment.
Solution Approach 2:
The AI processing unit is designed to work with multiple types of vehicles and can process various AI algorithms through software configuration rather than hardware redesign. This multi-functionality approach allows the same hardware platform to serve different applications and vehicle types, reducing overall system complexity.
3Device complexity
If simple rule-based algorithms are used in existing control device, then hardware can be kept low-end, but driver satisfaction and driving performance cannot be optimized
Solution Approach 1:
An AI processing unit acts as an intermediary between the simple rule-based legacy controller and the driver experience. This intermediary layer processes complex patterns from multiple sensors and generates optimized torque commands that enhance driver satisfaction and driving performance without requiring the legacy hardware to become complex.
Solution Approach 2:
The system transitions from one-dimensional rule-based control to multi-dimensional AI-driven control by incorporating numerous input variables (driver behavior patterns, environmental conditions, vehicle state) that were previously impossible to process. This dimensional expansion enables sophisticated optimization of driving performance while keeping the base hardware simple.
Data Source
AI summary
An embodiment vehicle driving control apparatus includes a processor configured to determine an output torque of a driving source by using a driver driving propensity for each driver of a plurality of drivers or a driving situation based on a learning algorithm and a memory configured to store algorithms and data to be driven by the processor.