Agricultural Sensor Fusion for Collision Avoidance

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Self-propelled agricultural working machines face challenges in achieving high operational efficiency while maintaining operational reliability due to high error detection rates from sensitive sensor arrangements, leading to unnecessary service interruptions and reduced efficiency, and the tendency to stop for imminent collisions that do not occur, reducing operating efficiency.

Innovation Solution

The integration of a sensor arrangement comprising multiple sensors that detect environmental information based on different physical properties, such as normal-light and thermal cameras, allows for differentiated object categorization and reaction, increasing operational safety and efficiency by generating detailed environmental information and assigning urgency levels to control actions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If the sensitivity of the sensor arrangement is increased to achieve high operational reliability, then collision detection capability is improved, but error detection rate increases leading to frequent unnecessary service interruptions and reduced operational efficiency

Engineering Contradiction:
Improveoperational reliabilityVSAvoidoperational efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The sensor arrangement is segmented into multiple independent sensors (first sensor and additional sensor) that detect different physical properties. This segmentation allows the system to process information from multiple sources and differentiate between true collision risks and false alarms, thereby maintaining high sensitivity while reducing unnecessary interruptions.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The driver assistance system acts as an intermediary that receives and processes sensor information from multiple sensors. It evaluates the combined information to determine whether a detected object represents a genuine collision risk, thereby filtering out false alarms while maintaining high detection sensitivity.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If the sensor arrangement stops the working machine for imminent collisions with high probability of not occurring, then safety is improved, but operating efficiency is considerably reduced

Engineering Contradiction:
ImprovesafetyVSAvoidoperating efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The sensor information from multiple sensors is segmented and evaluated separately before being combined. This allows the driver assistance system to assess the genuine risk level more accurately and avoid unnecessary stoppages for false collision risks.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system applies partial action by not stopping the machine for every detected object, but only for those that meet a higher threshold of confirmed risk based on multiple sensor corroboration. This reduces excessive stoppages while maintaining safety for genuine threats.

Inventive Principle:
Principle #16Partial or excessive action

3Measurement precision

If multiple sensors detecting different physical properties are integrated, then object categorization precision is improved, but device complexity increases

Engineering Contradiction:
Improveobject categorization precisionVSAvoidsensor arrangement complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The driver assistance system is designed with multi-functionality to process and evaluate information from multiple different sensor types. This universal processing capability allows the system to handle various sensor inputs without proportionally increasing overall system complexity.

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

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This approach enhances operational reliability and efficiency by enabling precise reactions to environmental objects, prioritizing critical situations, and optimizing control actions based on urgency levels, thereby reducing unnecessary interruptions and improving overall machine performance.

Implementation Method 1

the first sensor information is sensor information from a normal-light camera

Methodology Applied
Scientific EffectLight reflection: Reflection

Implementation Method 2

a thermal camera on the other. While the normal light camera provides information about the shape, the coloring or the like of the surrounding object, the thermal camera can be used to determine the temperature of the surrounding object

Methodology Applied
Scientific EffectThermal radiation: Thermal Radiation

Data Source

PatentEP3150055B1Self-propelled agricultural machine
Publication Date: 2020.02.19 CLAAS E SYSTEMS GMBH
  • EP3150055B1 patent drawingFigure 1
  • EP3150055B1 patent drawingFigure 2
  • EP3150055B1 patent drawingFigure 3

AI summary

The invention relates to a self-propelled agricultural work machine with at least one working element (3-8), in particular a drive system (3), and with a driver assistance system (9) for generating control actions within the work machine (1), wherein a sensor arrangement (10) is provided for generating environmental information (11-14), wherein the driver assistance system (9) generates the control actions based on the environmental information (11-14), wherein the sensor arrangement (10) has a first sensor (15) that detects a first sensor information (16-19) based on a first physical property of an environmental object (20, 21) in the environment of the work machine (1).It is proposed that the sensor arrangement (10) includes at least one further sensor (22) which acquires further sensor information (23-26) based on a further physical property of the environment object (20, 21) and that the driver assistance system (9) generates environment information (11-14) about the environment object (20, 21) from the first sensor information (16-19) and the further sensor information (23-26).