Fuzzy Driving Assistance for Speed and Lane Change Decisions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
There is a need for improved systems to facilitate autonomous driving, particularly for vehicles classified as level 2 or higher self-driving vehicles, which require advanced driving assistance to navigate and make decisions based on real-time environmental data.
Innovation Solution
The implementation of a fuzzy inference system in autonomous vehicles that uses sensors to obtain inputs such as distance and relative distance, applying fuzzy logic rules to generate outputs for speed control and lane change recommendations, utilizing Mamdani or Sugeno fuzzy inference systems with specific membership functions and rules to mimic human driving decisions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional binary logic control systems are used for autonomous driving, then the system structure is simple, but the system cannot handle complex driving scenarios and make nuanced decisions
Solution Approach 1:
The patent transforms the control system from binary logic (0 or 1) to fuzzy logic with continuous membership values between 0 and 1. This parameter change allows the system to represent partial truths and nuanced states, enabling it to handle complex driving scenarios with gradual transitions rather than abrupt binary decisions.
Solution Approach 2:
The patent introduces fuzzy membership functions as an intermediary layer between sensor inputs and control outputs. This intermediary allows for smooth transitions and nuanced decision-making by mapping crisp input values to fuzzy membership degrees, then combining multiple fuzzy rules to produce final control actions.
2Measurement precision
If fuzzy logic systems with multiple membership functions and rules are implemented, then decision-making accuracy improves, but computational complexity increases
Solution Approach 1:
The patent segments the fuzzy logic system into distinct modular components: input fuzzyification modules, separate fuzzy rule bases for different driving scenarios, and output defuzzification modules. This segmentation allows for manageable complexity by breaking down the overall system into smaller, independently configurable units that can be optimized separately.
Solution Approach 2:
The patent implements dynamic adjustment of fuzzy membership functions and rule priorities based on driving conditions. The system can adaptively modify membership function parameters and rule weights in real-time, allowing high precision when needed while reducing computational load during normal conditions through dynamic simplification.
3Reliability
If real-time sensor data processing with fuzzy inference is performed, then driving safety improves, but processing time increases
Solution Approach 1:
The patent pre-computes and stores fuzzy rule bases and membership function parameters during system initialization or offline training phases. By preparing these computational resources in advance, the real-time inference process only requires looking up pre-computed values and performing simple aggregation operations, significantly reducing real-time processing time while maintaining safety.
Solution Approach 2:
The patent implements a hierarchical fuzzy inference architecture that processes critical safety-related inputs through simplified fast-track fuzzy rules, while less time-sensitive inputs undergo more comprehensive processing. This allows the system to rush through essential safety decisions with reduced computational overhead while maintaining overall decision quality.
Data Source
AI summary
Advanced drying systems including a vehicle having a fuzzy inference system with a speed control logic set containing fuzzy logic rules, a lane change logic set containing fuzzy logic rules, or both, and related methods and systems are described.