Autonomous Vehicle Threat Detection and Relocation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for autonomous vehicles fail to effectively and autonomously handle threatening situations, such as persistent threats near parked vehicles, leading to potential safety risks and inefficiencies in threat recognition and response.

Innovation Solution

A method and device that utilize onboard sensors and cameras to detect the presence of threats, determine vehicle parking states, and autonomously drive the vehicle to a safe location, such as a well-lit area, while using AI-powered image recognition to differentiate between threats and authorized individuals, and alert authorities through audible and silent alarms.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If autonomous threat handling is implemented, then vehicle safety is improved, but system complexity increases

Engineering Contradiction:
Improvevehicle safetyVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The autonomous threat handling system is divided into distinct functional modules: sensor module for detecting motion and capturing images, processing module for analyzing images and determining threats, and response module for executing autonomous responses. This segmentation manages system complexity by organizing functions into separate, manageable components while maintaining overall safety effectiveness.

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If AI-powered image recognition is used to differentiate threats, then measurement precision is improved, but computing requirements and energy consumption increase

Engineering Contradiction:
Improvethreat recognition accuracyVSAvoidcomputing energy consumption
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The system applies AI-powered image recognition selectively rather than continuously - images are captured and analyzed only when motion is detected in the surrounding space. This partial action approach maintains high threat recognition accuracy when needed while significantly reducing computing energy consumption during normal parking periods without threats.

Inventive Principle:
Principle #16Partial or excessive action

Solution Approach 2:

The system performs preliminary motion detection using sensors before triggering the more computationally intensive AI image recognition. This preliminary action filters out unnecessary processing by only activating the energy-consuming AI recognition when motion is detected, thereby reducing overall energy consumption while maintaining precision when threats are present.

Inventive Principle:
Principle #10Preliminary action

3Reliability

If continuous monitoring is performed to detect persisting threats, then reliability is improved, but energy consumption increases

Engineering Contradiction:
Improvethreat detection reliabilityVSAvoidmonitoring energy consumption
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The system implements periodic monitoring by capturing images during specific time periods when motion is detected, rather than continuous monitoring. The system monitors for motion continuously but only activates image capture and analysis during detected motion events, creating a periodic action pattern that maintains threat detection reliability while significantly reducing energy consumption compared to continuous operation.

Inventive Principle:
Principle #19Periodic action

Data Source

PatentEP3561786B1Method and device for operating a vehicle
Publication Date: 2022.10.26 DR ING H C F PORSCHE AG
  • EP3561786B1 patent drawingFigure 1
  • EP3561786B1 patent drawingFigure 2

AI summary

A device and a method of operating a vehicle comprising detecting (101) a presence or an absence of an occupant inside the vehicle, determining (102) a vehicle parking state, detecting (103), a motion in a space surrounding the vehicle within a predetermined distance of the vehicle, capturing (104) at least one image of at least a part of the space surrounding the vehicle, detecting (105) from the at least one image a presence of a thread to the vehicle, determining (106) a reaction of the vehicle depending on the detected thread, starting (107) the determined response autonomously in case of absence of any occupant inside the vehicle and when the vehicle is in a vehicle parking state.