Agricultural Vehicle Field Work Detection via Trajectory Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for detecting and classifying agricultural field work rely on digital maps, which are not applicable to unrecorded fields, and lack accuracy in distinguishing between field work and transport activities.

Innovation Solution

A method using a satellite navigation receiver to record position fixes and time stamps, analyzing trajectory patterns, and determining feature vectors to detect field work without digital maps, incorporating maximum angle and distance criteria to identify field work, and classifying activities based on working width and speed.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If digital maps are used to detect fieldwork, then fieldwork detection can be performed on recorded fields, but the method cannot detect unrecorded fields and requires pre-existing map data

Engineering Contradiction:
Improvecapability to detect fieldworkVSAvoiddependency on digital maps
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent extracts the essential fieldwork detection capability from the digital map dependency by using satellite navigation position fixes and trajectory analysis alone. The method takes out the requirement for pre-recorded field boundaries and digital map data, enabling detection on any terrain including unrecorded fields through pure trajectory pattern recognition

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The system performs self-service by automatically detecting fieldwork and creating field boundaries through trajectory analysis without requiring external digital map data. The method uses the vehicle's own navigation data to identify working patterns, calculate field perimeters, and determine work areas independently of any pre-existing geographic information systems

Inventive Principle:
Principle #25Self-service

2Ease of operation

If simple trajectory recording is used, then the system is easy to operate, but accuracy in distinguishing fieldwork from transport activities is insufficient

Engineering Contradiction:
Improvesimplicity of system operationVSAvoidaccuracy of fieldwork detection
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The patent applies dynamics by analyzing the temporal and spatial characteristics of trajectory segments. The system dynamically evaluates angle relationships between consecutive trajectory segments, time intervals between position fixes, and geometric patterns to distinguish between transport movements and fieldwork operations, transforming static position data into dynamic behavioral recognition

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The method changes parameters by incorporating multiple evaluation criteria including angle thresholds, time interval ranges, trajectory segment lengths, and geometric shape characteristics. These parameter changes enable precise differentiation between fieldwork and transport activities while maintaining ease of operation through automated threshold-based classification

Inventive Principle:
Principle #35Parameter changes

3Measurement precision

If parallel driving patterns are analyzed with multiple criteria, then fieldwork detection accuracy is improved, but the computational complexity increases

Engineering Contradiction:
Improveaccuracy of fieldwork detectionVSAvoidcomputational processing requirements
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the trajectory into discrete trajectory segments between turning points, allowing independent analysis of each segment's angular relationships and geometric properties. This segmentation reduces computational complexity by breaking down continuous trajectory data into manageable discrete units that can be processed sequentially with simple angle and distance calculations

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP3449286B1Method for detecting agricultural field work performed by a vehicle
Publication Date: 2019.12.11 FARMDOK GMBH
  • EP3449286B1 patent drawingFigure 1
  • EP3449286B1 patent drawingFigure 2
  • EP3449286B1 patent drawingFigure 3

AI summary

The invention relates to a method for detecting agricultural field work performed by a vehicle (1), comprising: a) carrying a satellite navigation receiver (4) on the vehicle (1) and recording a sequence (FG) of position fixes (pi) and associated time stamps (ti); b) for a position fix (pi): determining a group (Gi) of position fixes that lie in a range (B); c) dividing the group (Gi) into sub-groups on the basis of a minimum time interval; d) determining trajectory sections (trj) on the basis of the sub-groups; e) determining an attribute vector (Mi) for the considered position fix (pi) comprising at least an angle (αi) between two trajectory sections (trj); and, f) if the attribute vector (Mi) satisfies a specified threshold value criterion (K), detecting the position fix (pi) as a location of field work (F).