Autonomous Maneuver Planning Around Parked Vehicles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles lack sufficient information to differentiate between temporarily and permanently parked vehicles, leading to potential extended stops behind stationary vehicles.
Innovation Solution
A system and method that classify remote stationary vehicles using kinematic and perception data, applying Kalman filters to determine probabilities of a vehicle being parked or non-parked, and calculating Von Neumann-Morgenstern cost values to determine actionable behaviors for the autonomous vehicle.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If the autonomous vehicle stops behind a stationary vehicle without additional classification information, then the vehicle can maintain a simple stopping behavior, but it may inadvertently stop behind a parked vehicle for an extended period losing time
Solution Approach 1:
The system performs preliminary classification of the stationary vehicle as either temporarily parked or permanently parked before determining the autonomous vehicle's behavior. This advance classification prevents extended unnecessary stops by identifying parked vehicles early, allowing the autonomous vehicle to plan appropriate maneuvers such as changing lanes or proceeding cautiously, thereby resolving the time loss without requiring complex real-time decision-making
Solution Approach 2:
The patent introduces an intermediary classification system that analyzes various data sources (sensor data, vehicle state information, environmental context) to determine the status of the stationary vehicle. This intermediary layer provides structured classification information that guides the autonomous vehicle's behavior selection, balancing the need for accurate vehicle status determination with system complexity by using a modular classification approach
2Measurement precision
If the autonomous vehicle uses multiple data sources and probability calculations to classify stationary vehicles, then the accuracy of vehicle status determination is improved, but the computational complexity increases
Solution Approach 1:
The classification system is segmented into distinct functional modules: data acquisition from multiple sensors, kinematic analysis, perception data processing, probability calculation, and behavior determination. Each module handles a specific aspect of the classification task, improving overall accuracy while managing complexity through modular design. This segmentation allows the system to process multiple data sources systematically without creating an unmanageably complex monolithic system
Solution Approach 2:
The system changes parameters by converting raw sensor data and vehicle state information into probability values that represent the likelihood of the stationary vehicle being temporarily or permanently parked. This parameter transformation enables the system to handle uncertainty in a quantitative manner, improving classification accuracy while maintaining computational tractability through probabilistic reasoning rather than requiring exhaustive analysis of all possible vehicle states
Data Source
AI summary
An autonomous vehicle, system and method of navigating the autonomous vehicle. The system includes one or more sensors for obtaining data with respect to a remote stationary vehicle, and a processor. The processor is configured to classify the remote stationary vehicle into an object hypothesis based on the data, determine an actionable behavior of the autonomous vehicle based on a probability for the object hypothesis, and navigate the autonomous vehicle with respect to the remote vehicle via the actionable behavior.