Autonomous Vehicle Behavior Data Mining for Lane-Level Driving Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Ensuring the safety of autonomous driving systems by improving the processing and utilization of behavior data to enhance decision-making in autonomous vehicles.
Innovation Solution
A method for processing behavior data that includes acquiring historical driving data, performing data mining to extract features such as lane-change positions, traveling speeds, and path features, and using these features along with perceptual positioning information to control autonomous vehicles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If basic technologies such as machine vision, radar positioning, satellite positioning and intelligent control are used to implement autonomous driving function, then the autonomous driving function can be achieved, but the safety of autonomous driving cannot be ensured and continuously improved
Solution Approach 1:
The patent applies preliminary action by collecting and storing historical driving behavior data before actual driving decisions are needed. The system pre-processes behavior data from multiple vehicles to extract driving feature information in advance, creating a knowledge base that can be quickly accessed during autonomous driving operations. This allows the system to improve safety without adding complex real-time processing requirements.
Solution Approach 2:
The patent introduces driving feature information as an intermediary between raw behavior data and driving decisions. The system extracts meaningful features such as lane-change positions, traveling speeds, and path characteristics from raw data, then uses these processed features to inform autonomous driving decisions. This intermediary layer transforms unstructured behavior data into actionable insights that improve safety.
2Measurement precision
If more behavior data is collected and processed, then the decision-making accuracy improves, but the data processing complexity and time consumption increase
Solution Approach 1:
The patent applies extraction by selectively extracting only the most relevant driving feature information from large volumes of behavior data. Instead of processing all raw data, the system identifies and extracts key features such as lane-change positions, traveling speeds, and path characteristics. This extraction approach maintains high decision-making accuracy while significantly reducing processing complexity.
Solution Approach 2:
The patent transforms raw behavior data into standardized driving feature parameters through data mining and processing. By converting diverse raw data into consistent parameter formats (position coordinates, speed values, path descriptors), the system enables efficient comparison and analysis without requiring complex processing of heterogeneous data structures.
3Reliability
If lane-level navigation data is included in historical driving data, then the driving feature information becomes more comprehensive, but the data storage and processing requirements increase
Solution Approach 1:
The patent uses copying by storing processed driving feature information that replicates the essential characteristics of comprehensive lane-level navigation data. Instead of storing all raw navigation data, the system creates condensed copies in the form of extracted features (lane-change positions, path characteristics) that capture the necessary information for safe autonomous driving decisions with reduced storage requirements.
Data Source
AI summary
A method for processing behavior data, a method for controlling an autonomous vehicle, apparatuses thereof, a device, a storage medium, a computer program product, and an autonomous vehicle are provided. The method includes: acquiring historical driving data, the historical driving data comprising lane-level navigation data; and performing data mining on the historical driving data to obtain driving feature information, the driving feature information comprising at least one of: a lane-change position feature, a traveling speed feature, or a traveling path feature.