Driving Assistance Correction Using Position-Based Feature Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing driving assistance systems face issues with uncomfortable driving experiences due to mismatched vehicle speed and steering positions based on map information, leading to high memory usage and processing loads, and unintended assistance during unexpected events.
Innovation Solution
A driving assistance device that switches between automatic and manual driving modes, using an information receiver, determination processor, storage, correction processor, and vehicle controller to acquire and correct driving operations based on driver inputs, reducing memory usage and processing load while matching driving assistance to the driver's preferences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If driving operations are learned based on driver behavior to provide personalized assistance, then driving assistance comfort is improved, but memory usage and processing load increase significantly
Solution Approach 1:
The patent extracts only the essential feature amounts from driving operations (accelerator pedal operation amount, brake pedal operation amount, steering wheel operation amount) and stores them as correction amounts associated with specific position information, rather than storing complete driving operation data. This selective extraction reduces memory usage while preserving the ability to provide personalized driving assistance.
Solution Approach 2:
The patent stores correction amounts locally associated with specific position information (GPS coordinates, map information) rather than maintaining a global database of all driving operations. This allows the system to provide personalized assistance at specific locations without storing comprehensive driving data everywhere, thereby reducing overall memory requirements.
2Measurement precision
If comprehensive driving operation data is stored to learn optimal driving behavior, then driving assistance accuracy is improved, but system processing load increases
Solution Approach 1:
The system extracts only the necessary feature amounts (operation amounts of accelerator, brake, and steering wheel) from comprehensive driving operation data and stores these as correction amounts. This extraction approach maintains driving assistance accuracy by capturing essential behavioral patterns while significantly reducing processing load by avoiding storage and analysis of unnecessary detailed operation data.
Solution Approach 2:
The patent transforms comprehensive driving operation data into simplified correction amount parameters that can be directly applied to automatic driving control. By changing the parameter representation from detailed operation sequences to concise correction values, the system achieves accurate driving assistance with reduced computational complexity.
3Adaptability or versatility
If all driving operations in target areas are recorded as feature points, then driving assistance coverage is improved, but unintended assistance during unexpected events occurs
Solution Approach 1:
The system uses feedback from driver operations to determine whether to store correction amounts. When the driver manually operates the vehicle, the system learns from these operations and stores correction amounts for future automatic driving. However, the system can incorporate feedback mechanisms to identify and exclude abnormal operations (such as emergency braking or sudden steering to avoid obstacles) from being stored, thereby preventing unintended assistance while maintaining broad coverage.
4Manufacturing precision
If unique speed and steering positions are determined from map information, then automatic driving control precision is improved, but driver comfort deteriorates when mismatched with driver intent
Solution Approach 1:
The patent merges map information-based automatic driving control with driver behavior-based correction amounts. The system determines speed and steering positions from map information to maintain control precision, then combines these with stored correction amounts derived from actual driver operations. This merging allows the system to maintain precise automatic control while adapting to driver preferences and intentions, thereby improving driver comfort.
Solution Approach 2:
The system dynamically adjusts control parameters by applying correction amounts to the base values derived from map information. This parameter modification allows the automatic driving control to deviate from strictly map-based values when necessary to match driver intent, improving comfort while maintaining the precision framework provided by map data.
Data Source
AI summary
Position information of a vehicle 100 and feature amounts of driving operations by a driver are acquired as triggered by occurrence of switching from an automatic driving mode to a manual driving mode, and a driving operation to be corrected in the automatic driving mode and a correction amount thereof, are determined from these feature amounts. A driving operation in the automatic driving mode is corrected using the determination result, so that the vehicle 100 is controlled in the automatic driving mode in which the corrected driving operation is included.