In-Vehicle Object Search Feedback Using Speech and Semantic Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing vehicle systems that use robotic arms for object searching occupy valuable space and distract drivers, increasing the risk of accidents by diverting attention from the road.

Innovation Solution

A feedback method and apparatus that identifies the location of a target object within a vehicle environment using image capture and semantic segmentation, converting the information into speech for the driver without requiring visual attention.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If a robotic arm is used to grasp and search for lost objects, then the user's object-searching demand is satisfied, but the vehicle cabin space is substantially occupied and the user's driving and riding experience is seriously affected

Engineering Contradiction:
Improveobject-searching capabilityVSAvoidvehicle cabin space
Core Design Contradiction:
Ease of operationVSArea of stationary object

Solution Approach 1:

The patent replaces the mechanical robotic arm system with an optical sensing system (camera) and information processing system (semantic segmentation algorithm, speech synthesis). Instead of using a mechanical device to physically search for objects, the system captures images, processes them through semantic segmentation to identify object locations, and provides feedback through speech, thereby eliminating the need for physical space occupation while maintaining object-searching functionality

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent introduces an intermediary information processing system that mediates between the object-searching demand and the driver. Rather than directly interacting with objects through a robotic arm, the system uses image capture, semantic analysis, and speech feedback as intermediaries to provide indirect but effective object location information, avoiding direct mechanical intervention in the cabin space

Inventive Principle:
Principle #24Intermediary (Mediator)

2Loss of information

If the driver transfers attention to search for an object, then the object can be located, but the driver cannot focus on driving and the response time upon confronting an accident event is prolonged

Engineering Contradiction:
Improveobject location informationVSAvoiddriving safety
Core Design Contradiction:
Loss of informationVSReliability

Solution Approach 1:

The system enables the driver to maintain attention on driving while the system autonomously performs the object-searching task. The driver simply issues a voice command and receives automated speech feedback about object locations, allowing the system to serve itself in finding objects without requiring the driver's visual attention or cognitive resources, thereby maintaining driving safety

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent implements a feedback mechanism where the system captures images, processes them through semantic segmentation to determine object locations, and provides real-time speech feedback to the driver. This closed-loop feedback system delivers object location information without requiring the driver to visually search, maintaining continuous attention on the road while providing accurate object location data

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP4708227A1Feedback method and apparatus for object searching in vehicles
Publication Date: 2026.03.11 CARIAD (CHINA) CO LTD
  • EP4708227A1 patent drawingFigure 1~2
  • EP4708227A1 patent drawingFigure 3~4
  • EP4708227A1 patent drawingFigure 5

AI summary

The present invention provides a feedback method (1000) and apparatus for object searching in a vehicle. The method comprises identifying (S1100) information about vehicle information where a target object is located, based on an object position of the target object; and converting (S1200) the information about the vehicle environment into speech information. According to the present invention, it is possible to identify environment information of the target object based on the object position of the target object and convert the environment information into speech information for feedback, thereby effectively satisfying the user's object searching demand and improving the user's experience without affecting the user's driving and riding space.