Human Driving Rule Models for Autonomous Road Decisions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in making real-time driving decisions, particularly when encountering uncontrolled traffic intersections, stationary vehicles, and jaywalkers, requiring significant processing power and network connectivity, which can be resource-intensive and inefficient.
Innovation Solution
The development of models based on human driving behavior, using neural networks and data analysis to generate rules for autonomous vehicle maneuvers, such as merging at intersections, passing stationary vehicles, and avoiding jaywalkers, allowing for reduced processing power and network connectivity by pre-defining decision criteria based on human driving patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If autonomous vehicle systems use sensor data to make driving decisions in real-time, then the vehicle can respond to dynamic environmental changes, but the processing power and network connectivity requirements increase significantly
Solution Approach 1:
The system pre-processes sensor data and pre-determines driving decisions for various traffic scenarios before the vehicle encounters them. By performing computations in advance and storing decision rules, the system reduces real-time processing requirements while maintaining responsive decision-making capability.
Solution Approach 2:
The patent divides complex driving scenarios into distinct traffic situations (e.g., uncontrolled intersections, stationary vehicles, jaywalkers) and creates separate decision models for each. This segmentation allows the system to handle specific scenarios efficiently without processing all possible situations in real-time, reducing overall computational burden.
2Reliability
If autonomous vehicle systems process complex driving scenarios in real-time, then decision accuracy improves, but the system complexity and computational requirements increase
Solution Approach 1:
The system pre-determines driving decisions for various traffic scenarios and stores them as decision rules. By performing complex analysis beforehand and caching results, the system maintains high decision accuracy without requiring complex real-time processing infrastructure.
Solution Approach 2:
The patent creates simplified decision models that replicate human driving behavior patterns for specific scenarios. These models copy essential human judgment logic without requiring full human-level cognitive processing, reducing system complexity while maintaining reliable decision-making.
3Reliability
If autonomous vehicle systems use human driving behavior data to improve decision making, then the reliability and safety increase, but the data processing and model generation complexity increase
Solution Approach 1:
The system segments human driving behavior data into scenario-specific patterns (uncontrolled intersections, stationary vehicles, jaywalkers) and creates dedicated decision models for each. This segmentation simplifies the overall modeling process by breaking down complex human behavior analysis into manageable, scenario-specific components.
4Use of energy by moving object
If autonomous vehicle systems pre-define decision criteria based on human driving patterns, then processing power requirements reduce, but the adaptability to new scenarios decreases
Solution Approach 1:
The patent creates a universal decision-making framework that can handle multiple traffic scenarios using a common set of pre-determined rules and models. This framework is designed to be extended to new scenarios by adding appropriate decision rules, maintaining adaptability while keeping computational requirements low through reuse of the core decision engine.
Data Source
AI summary
The current disclosure provides techniques for using human driving behavior to assist in decision making of an autonomous vehicle as the autonomous vehicle encounters various scenarios on the road. For each scenario, a model may be generated based on human driving behavior that governs how an autonomous vehicle maneuvers in that scenario. As a result of using these models, reliability and safety of autonomous vehicle may be improved. In addition, because the model is programmed into the autonomous vehicle, the autonomous vehicle, in many instances, need not consume resources to implement complex calculations to determine driving behavior in real-time.