Autonomous Lane Change Control With Neural Network Rule Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing lane change decision-making systems for autonomous driving face challenges in complex environments, as they struggle to exhaustively consider all possible scenarios, leading to inefficiencies and safety concerns due to overly complex rules or limitations in machine learning models.
Innovation Solution
A method that combines neural network processing with predetermined rules, utilizing velocity difference suppression, fast lane priority, and decision cooling rules to generate an intelligent lane change strategy, incorporating velocity and lane information, and state information of adjacent vehicles, to optimize lane change decisions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If machine learning models are used for lane change decisions, then adaptability to complex scenarios is improved, but decision reliability deteriorates due to model limitations
Solution Approach 1:
The patent merges machine learning models with rule-based systems to create a hybrid decision-making architecture. The neural network processes complex scenario features while predetermined rules provide reliable decision boundaries, combining the adaptability of ML with the reliability of rule-based approaches.
Solution Approach 2:
The patent introduces an intermediary rule-based layer that mediates between the neural network output and final lane change decisions. This intermediary layer applies predetermined rules to adjust or override neural network predictions, ensuring decision reliability while maintaining adaptability.
2Reliability
If rule-based systems are used for lane change decisions, then decision reliability is improved, but adaptability to complex scenarios deteriorates due to inability to exhaust all scenarios
Solution Approach 1:
The patent combines rule-based systems with machine learning models to create a hybrid architecture where rules provide reliable decision frameworks and the neural network handles complex scenario variations that rules cannot exhaustively cover.
Solution Approach 2:
The patent applies partial rule-based control where predetermined rules handle specific critical scenarios while the neural network handles the remaining complex scenarios, recognizing that exhaustive rule coverage is impossible.
3Adaptability or versatility
If complex rules are applied to cover all driving scenarios, then adaptability is improved, but system complexity deteriorates making implementation difficult
Solution Approach 1:
The patent segments the decision-making system into distinct modules: a neural network for feature processing and a separate rule-based layer for decision adjustment. This segmentation allows each module to handle specific aspects independently, reducing overall system complexity while maintaining comprehensive scenario coverage.
Solution Approach 2:
The patent introduces a simplified intermediary rule layer that acts as a mediator between the neural network and final decisions. This intermediary uses predetermined rules to adjust neural network outputs without requiring complex rule sets, reducing implementation difficulty while maintaining adaptability.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The disclosure relates to a lane change method and system, a storage medium, and a vehicle. The lane change method includes the following steps: receiving consecutive frames of condition information, the condition information including velocity information of a current vehicle, state information of an adjacent vehicle, and lane information; with the condition information as an input to a neural network, processing the condition information by means of the neural network, to obtain an initial lane change strategy; and correcting the initial lane change strategy based on a predetermined rule and the condition information, to generate and output a corrected lane change strategy. According to this lane change method, intelligent, safe and efficient lane change may be achieved during an autonomous driving or driving assistance process.