Vehicle Object Detection with Pose Prediction Beyond View Angle

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing object detection systems face reduced tracking accuracy when the object moves out of the capture view angle, affecting distance, position, and pose detection.

Innovation Solution

An object detection device comprising a distance detection portion, position detection portion, pose detection portion, and prediction portions that utilize stereo image capture and intervehicle communication to accurately detect and predict the position and pose of objects regardless of view angle or distance, switching between actual detection and prediction results based on object visibility.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If image analysis is used to detect orientation and speed of preceding vehicle, then tracking accuracy is improved when object is within view angle, but tracking accuracy deteriorates when object moves out of capture view angle

Engineering Contradiction:
Improvetracking accuracyVSAvoiddetection capability outside view angle
Core Design Contradiction:
Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The system performs preliminary action by predicting the future position and pose of the preceding vehicle using detected state information (orientation, speed) before the vehicle actually moves out of the capture view angle. This prediction enables the system to maintain tracking accuracy even when the object is no longer visible in the current frame.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent introduces an intermediary mechanism by using state information (orientation, speed, position) as a mediator between direct image analysis and position prediction. When the object moves out of view, this intermediary state information allows the system to continue tracking through prediction rather than direct observation.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Device complexity

If only image capture device is used for detection, then device complexity is reduced, but detection reliability deteriorates when object is at various distances and view angles

Engineering Contradiction:
Improvedetection system complexityVSAvoiddetection reliability
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The patent applies universality by designing a multi-functional detection system that can operate in multiple modes: direct image analysis when the object is within view angle, and prediction-based tracking when the object moves out of view. This universal system maintains high detection reliability across various distances and view angles without requiring multiple specialized devices.

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

Data Source

PatentUS12033395B2Object detection device, travel control system, and travel control method
Publication Date: 2024.07.09 ASTEMO LTD
  • US12033395B2 patent drawing
  • US12033395B2 patent drawing
  • US12033395B2 patent drawing

AI summary

A problem of the present invention is to provide an object detection device etc. that can accurately detect an object regardless of a view angle position of and distance to the object. An object detection device of the present invention has: a stereo distance detection portion 105 that detects a distance to an object; a position detection portion 106 that detects a position of the object; a pose detection portion 111 that detects a pose of the object; a vehicle information input portion that inputs state information about a host vehicle and a different vehicle; a position prediction portion 109 that predicts a position of the different vehicle based on the state information about the host vehicle and the different vehicle; a pose prediction portion 110 that predicts a pose of the different vehicle based on the state information about the host vehicle and the different vehicle; and a determination portion 112 that determines a distance to, a position of, and a pose of the different vehicle in response to the information detected or predicted by the distance detection portion, the position detection portion, the pose detection portion, the position prediction portion, and the pose prediction portion.