Autonomous Vehicle Navigation Using Parked Vehicle State Prediction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Autonomous vehicles face challenges in navigating safely and accurately due to slow reaction times of traditional algorithms, failure to utilize environmental cues like parked car directions and pedestrian emergence indicators, and lack of integration of visual and infrared data for comprehensive situational awareness.

Innovation Solution

A system that includes cameras and processing units analyzing visual and infrared data to determine the state of parked vehicles, predict their movement, and adjust navigation paths, integrating data from various sensors and maps for enhanced situational awareness and decision-making.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If traditional algorithms are used for autonomous braking and navigation, then the system is simpler to implement, but the reaction time is too slow to match human drivers

Engineering Contradiction:
Improvereaction timeVSAvoidalgorithm complexity
Core Design Contradiction:
SpeedVSDevice complexity

Solution Approach 1:

The system performs preliminary classification of detected objects into vehicle types (parked, moving, emergency) and predicts their future states before navigation decisions are required. This advance processing enables faster reaction times when actual navigation decisions must be made.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The algorithm dynamically adjusts its processing based on detected object characteristics, using different levels of analysis for different vehicle types. Emergency vehicles receive immediate high-priority processing, while parked vehicles use predictive modeling, optimizing the balance between speed and accuracy.

Inventive Principle:
Principle #15Dynamics

2Measurement precision

If traditional single-sensor systems are used, then the device complexity is lower, but the situational awareness and measurement accuracy are insufficient

Engineering Contradiction:
Improveenvironmental detection accuracyVSAvoidsensor integration complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system merges visual camera data with infrared sensor data to create a comprehensive environmental model. This multi-sensory integration allows the vehicle to detect objects and characteristics that would be invisible to either sensor type alone, significantly improving measurement precision.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The integrated sensor system serves multiple functions simultaneously: visual cameras detect vehicle types and road conditions during daytime, while infrared sensors detect thermal signatures of pedestrians and vehicles during nighttime or adverse conditions, creating a universal detection system that operates effectively in all environments.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Loss of information

If parked vehicle characteristics are not analyzed, then the processing load is reduced, but important environmental cues like one-way road indicators and pedestrian emergence risks are missed

Engineering Contradiction:
Improveenvironmental information completenessVSAvoidprocessing efficiency
Core Design Contradiction:
Loss of informationVSProductivity

Solution Approach 1:

The system extracts specific relevant characteristics from parked vehicles (orientation, spacing patterns, position relative to road geometry) rather than processing all possible vehicle data. This selective extraction captures critical environmental information while maintaining processing efficiency.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The system uses the characteristics of detected parked vehicles to create predictive models of road geometry and pedestrian behavior patterns. These models serve as virtual copies of environmental conditions, allowing the system to infer road type and pedestrian risks without direct observation of every potential hazard.

Inventive Principle:
Principle #26Copying

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

The system enables autonomous vehicles to react promptly to changing circumstances, improve navigation accuracy, and enhance safety by predicting parked vehicle movements and integrating multi-sensory data for comprehensive environmental analysis.

Implementation Method 1

autonomous vehicle systems may employ infrared cameras to assess the environment and make predictions

Methodology Applied
Scientific EffectInfrared radiation detection: Infrared Radiation

Data Source

PatentEP3614105B1Controlling host vehicle based on detected parked vehicle characteristics
Publication Date: 2023.10.18 MOBILEYE VISION TECH LTD
  • EP3614105B1 patent drawingFigure 1
  • EP3614105B1 patent drawingFigure 2A
  • EP3614105B1 patent drawingFigure 2B

AI summary

Systems and methods are provided for navigating an autonomous vehicle. One embodiment relates to systems and methods for navigating a host vehicle based on detecting a door opening event. Another embodiment relates to systems and methods for navigating a host vehicle based on movement of a target vehicle toward a lane being traveled by the host vehicle. A third embodiment relates to systems and methods for detecting whether a road on which a host vehicle travels is a one-way road. A fourth embodiment relates to systems and methods for determining a predicted state of a parked vehicle in an environment of a host vehicle. A fifth embodiment relates to systems and methods for navigating a host vehicle based on a spacing between two stationary vehicles.