Object Recognition Driving Pattern Server for Autonomous Vehicle Control

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional autonomous driving systems struggle to adapt vehicle control strategies based on the type and risk of unrecognized objects, leading to limited situational awareness and stability in vehicle behavior control.

Innovation Solution

An object recognition-based driving pattern managing server that recognizes objects using big data, determines driving patterns based on object type and attributes, and transmits priority information for vehicle behavior scenarios to improve control strategies, enhancing the recognition success rate and stability of autonomous driving.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If the autonomous driving system recognizes only specific objects (vehicle, pedestrian, bicycle, truck, motorcycle), then the vehicle control strategy can be determined for recognized objects, but the system cannot adapt control strategies for unrecognized objects leading to limited situational awareness

Engineering Contradiction:

Improvevehicle control strategy determinationVSAvoidsituational awareness

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

The server system provides universal object recognition capabilities for diverse object types beyond predefined categories. By using image recognition technology and pattern matching against stored object data, the system can identify various objects (animals, obstacles, unusual vehicles) and provide appropriate driving pattern recommendations, making the autonomous driving system adaptable to unrecognized objects while maintaining reliable control strategy determination

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

A server acts as an intermediary between the autonomous driving system and object recognition. The server stores comprehensive object pattern data and processes image information to identify objects that the vehicle's onboard system cannot recognize. This intermediary enables the vehicle to access expanded object knowledge without modifying its core recognition hardware, thereby improving situational awareness while maintaining system reliability

Inventive Principle:

Principle #24Intermediary (Mediator)

2Device complexity

If the autonomous driving system uses conventional object recognition methods, then the system structure remains simple, but the driving strategy cannot be changed depending on the risk of the situation according to the type of object

Engineering Contradiction:

Improvesystem structureVSAvoiddriving strategy adaptation

Core Design Contradiction:

Device complexityVSAdaptability or versatility

Solution Approach 1:

The system adds a cloud-based server dimension to the autonomous driving architecture. Instead of expanding onboard recognition hardware, the solution moves object recognition and pattern matching to a remote server that processes image data and returns driving pattern recommendations. This dimensional shift enables sophisticated driving strategy adaptation based on object type and risk assessment without increasing vehicle system complexity

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

The patent replaces complex onboard mechanical recognition systems with software-based image recognition and pattern matching performed on a remote server. By substituting physical recognition hardware with computational image analysis, the system achieves enhanced object identification and driving strategy adaptation capabilities while maintaining simple vehicle system architecture

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Speed

If the system recognizes only predefined objects, then the recognition process is fast and simple, but the recognition success rate for diverse objects is limited

Engineering Contradiction:

Improverecognition speedVSAvoidrecognition success rate

Core Design Contradiction:

SpeedVSMeasurement precision

Solution Approach 1:

The server pre-stores comprehensive object pattern data including images, characteristics, and identification features for various object types during system setup. This preliminary preparation of recognition data enables the system to quickly compare incoming image information against the pre-loaded pattern database, maintaining fast recognition speed while expanding the ability to identify diverse objects including animals, obstacles, and unusual vehicles

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS11699292B2Apparatus for managing driving pattern based on object recognition, vehicle driving controlling apparatus using the same and method thereof

Publication Date: 2023.07.11 HYUNDAI MOTOR CO LTD

AI summary

An object recognition-based driving pattern managing server, a driving control apparatus of a vehicle using the same, and a method are provided. A server includes a processor recognizing a target object extracted from image information and determining a priority for each vehicle behavior scenario based on an accident rate for each vehicle behavior scenario among patterns similar to a mother pattern of the recognized target object to transmit the priority for each vehicle behavior scenario to a vehicle and a storage storing at least one or more of information about an image similar to the target object, target object information, pattern information for each target object, vehicle behavior scenario information, and priority information for each vehicle behavior scenario based on the target object.