Autonomous Lawnmower Self-Localization via Image Comparison

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing autonomous locomotion devices, such as lawnmowers, face challenges in precise self-localization and movement strategy development, especially in dynamic environments with prominent visual objects, requiring significant computing power and lacking independence from specific visual features.

Innovation Solution

The device employs an image acquisition unit, computing unit, and memory unit to compare current environmental images with stored sequences, generating optical flow and motion vectors to develop a movement strategy, allowing for precise self-localization and autonomous navigation independent of prominent visual objects, using a method that requires minimal computing power and accounts for historical data during a learning process.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If autonomous locomotion devices use prominent visual objects for self-localization, then localization can be achieved, but the device becomes dependent on specific visual features and requires significant computing power

Engineering Contradiction:
Improveself-localization precisionVSAvoidcomputing power requirement
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system performs preliminary actions by storing multiple environmental images and sequences in advance during a learning phase. These pre-stored images serve as reference data for later self-localization, eliminating the need for complex real-time processing of prominent visual objects.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system creates copies of environmental images and stores them in memory units. Instead of processing complex visual features in real-time, the device compares current sensor data with stored image copies to achieve self-localization, significantly reducing computing power requirements.

Inventive Principle:
Principle #26Copying

2Ease of operation

If autonomous locomotion devices rely on prominent visual objects for navigation, then movement strategy can be developed, but the device lacks independence from specific visual features

Engineering Contradiction:
Improvemovement strategy developmentVSAvoidindependence from visual features
Core Design Contradiction:
Ease of operationVSAdaptability or versatility

Solution Approach 1:

The system uses environmental images that serve multiple functions: they are stored as reference data, used for self-localization, and serve as basis for movement strategy development. This multi-functional approach eliminates dependence on specific prominent visual objects while maintaining ease of operation.

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

Solution Approach 2:

The system changes the parameter basis for navigation from relying on prominent visual objects to using stored environmental image sequences. This parameter change enables the device to develop movement strategies independently of specific visual features while maintaining operational ease.

Inventive Principle:
Principle #35Parameter changes

3Loss of time

If autonomous locomotion devices process environmental images in real-time, then current position can be determined, but computing power consumption increases

Engineering Contradiction:
Improveposition determination timeVSAvoidcomputing power consumption
Core Design Contradiction:
Loss of timeVSUse of energy by moving object

Solution Approach 1:

Environmental images are stored in advance during a learning phase, creating a reference library before actual navigation begins. This preliminary action eliminates the need for complex real-time processing, reducing both time loss and computing power consumption during operation.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system stores copies of environmental images in memory units for later comparison. This copying approach allows rapid position determination through simple comparison operations rather than complex real-time processing, significantly reducing computing power consumption.

Inventive Principle:
Principle #26Copying

Data Source

PatentEP2922384B1Autonomous locomotion device
Publication Date: 2018.02.14 ROBERT BOSCH GMBH
  • EP2922384B1 patent drawingFigure 1

AI summary

The invention relates to an autonomous transportation device, in particular to an autonomous lawnmower, having at least one image capture unit (12) for recording at least one environmental image and/or at least one environmental image sequence, having at least one computation unit (14) at least for evaluating the recorded environmental image and/or the environmental image sequence and having at least one memory unit (16) for storing at least one environmental image and/or at least one environmental image sequence. According to the invention the computation unit (14), in at least one operating state for creating a movement strategy, executes a movement estimation by comparing at least one environmental image recorded by the image capture unit (12) and/or an environmental image sequence recorded by the image capture unit (12) with an environmental image stored in the memory unit (16) and/or with an environmental image sequence stored in the memory unit (16).