Mobile Robot Context Analysis for Multi-Object Collision Avoidance

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current electronic devices, such as robots, lack the ability to accurately analyze and predict movements of objects in their environment due to the failure to consider relationships between multiple objects, leading to potential collisions and inaccurate situation analysis.

Innovation Solution

An electronic apparatus equipped with a camera, memory, and processor that identifies attribute information and environment information to determine control operations, including the relationship between objects, predicting movements and adjusting its traveling state to avoid collisions and maintain appropriate modes based on context analysis.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If the electronic apparatus considers relationships between multiple objects and their attributes, then the accuracy of situation analysis and prediction is improved, but the computational complexity and processing time increase

Engineering Contradiction:
Improveaccuracy of situation analysisVSAvoidcomputational complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The processor segments the analysis by first identifying individual objects and their attributes separately, then analyzing relationships between specific pairs of objects. This divides the complex multi-object analysis into manageable sequential steps, reducing computational complexity while maintaining accuracy.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system performs preliminary identification of objects and their attributes before analyzing relationships. By pre-processing and storing object attributes in structured formats, the system prepares data in advance for relationship analysis, reducing real-time computational burden while improving analysis accuracy.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If the electronic apparatus identifies and predicts movements of multiple objects based on relationships, then collision avoidance capability is improved, but the processing time and computational resources increase

Engineering Contradiction:
Improvecollision avoidance capabilityVSAvoidprocessing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The processor performs preliminary identification of object attributes and relationships before movement prediction. By pre-analyzing static attributes and relationship patterns, the system reduces real-time processing requirements while maintaining accurate collision avoidance predictions.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system prioritizes processing by focusing computational resources on identifying critical relationships and predicted movements that pose collision risks, skipping or simplifying analysis of non-critical object interactions. This rushes through essential analysis quickly while maintaining safety.

Inventive Principle:
Principle #21Skipping (Rushing through)

3Measurement precision

If the electronic apparatus analyzes relationships between dynamic and static objects, then the accuracy of predicted movement is improved, but the complexity of object identification and classification increases

Engineering Contradiction:
Improveaccuracy of predicted movementVSAvoidcomplexity of object identification
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The processor segments objects into dynamic and static categories based on identified attributes, then applies relationship analysis specifically to relevant object pairs. This segmentation simplifies the identification process by creating clear classification criteria while maintaining accurate movement prediction for dynamic objects.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP3715994B1Electronic apparatus and control method thereof
Publication Date: 2022.09.21 SAMSUNG ELECTRONICS CO LTD
  • EP3715994B1 patent drawingFigure 1
  • EP3715994B1 patent drawingFigure 2
  • EP3715994B1 patent drawingFigure 3

AI summary

An electronic apparatus includes a camera; a memory configured to store attribute information and environment information; a processor configured to identify a plurality of objects based on an image obtained by the camera, to identify a first context of a first object, from among the plurality of objects, based on a relationship between attribute information of the plurality of objects and the environment information, and to control a traveling state of the electronic apparatus based on the first context.