Rear Collision Warning Using Bounding-Volume Risk Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Teleoperated vehicles face a high risk of rear-end collisions due to violations of traffic rules by other participants, necessitating improved control methods to prevent such collisions.

Innovation Solution

A method involving computer vision techniques to identify rear risk objects through bounding volumes in rear-view images, determining distance information, and performing safety maneuvers to avoid collisions, utilizing video and audio data from sensors and real-time object detection systems.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If teleoperation is used to remotely control vehicles, then vehicle mobility service can be provided without drivers inside vehicles, but the risk of rear-end collisions increases due to delayed reaction time and reliance on remote operators

Engineering Contradiction:
Improvevehicle operation without internal driverVSAvoidcollision prevention capability
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The patent introduces an automated detection and warning system as an intermediary between the remote environment and the teleoperated vehicle. The system includes sensors (cameras, LIDAR, radar) that detect objects in the vehicle's path, process this data locally, and provide real-time warnings to both the remote operator and the vehicle's control system, creating a mediating layer that enhances safety without requiring the operator to be physically present

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system performs preliminary detection and analysis of potential collision risks before they materialize. By continuously monitoring the environment with sensors and processing data in real-time, the system identifies hazards in advance and provides early warnings to the teleoperator, enabling preventive action before the situation escalates to a actual collision

Inventive Principle:
Principle #10Preliminary action

2Extent of automation

If real-time remote monitoring and control is implemented, then vehicle teleoperation is enabled, but system complexity increases due to required sensors, communication infrastructure, and processing systems

Engineering Contradiction:
Improveremote vehicle control capabilityVSAvoidsystem infrastructure requirements
Core Design Contradiction:
Extent of automationVSDevice complexity

Solution Approach 1:

The patent employs multi-functional sensor systems that serve multiple purposes: cameras detect visual information for both monitoring and collision detection, LIDAR provides both environmental mapping and distance measurement, and radar detects both object presence and velocity. This multi-functionality reduces the need for separate specialized systems, thereby managing complexity while maintaining comprehensive teleoperation capabilities

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

Solution Approach 2:

The system includes onboard processing capabilities that automatically analyze sensor data and generate warnings without requiring constant remote intervention. The teleoperated vehicle's system processes local environmental data, identifies risks, and provides alerts autonomously, reducing the communication burden and simplifying the overall system architecture by distributing intelligence between vehicle and remote operator

Inventive Principle:
Principle #25Self-service

Data Source

PatentEP4273654A1Rear collision warning for vehicles
Publication Date: 2023.11.08 VAY TECH GMBH
  • EP4273654A1 patent drawingFigure 1A
  • EP4273654A1 patent drawingFigure 1B
  • EP4273654A1 patent drawingFigure 2A

AI summary

The present invention relates to a method for controlling a vehicle, in particular a teleoperated vehicle, comprising the steps of: providing an image of a rear view of the vehicle; performing a computer vision technique on the image, thereby identifying a rear risk object with a bounding volume; determining an information, in particular a distance information, of the rear risk object based on at least one parameter of the bounding volume; and performing a maneuver with the vehicle based on the information, in particular the distance information.