Dynamic Road Detection Using Machine Telemetry

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing fleet management systems struggle to accurately detect and maintain the changing travel routes of machines in dynamic work environments, such as construction or mining sites, due to the limitations of pre-defined routes and reliance on static virtual models.

Innovation Solution

A method and system that utilizes telemetry data from multiple machines to identify similar trajectories as roads, using a route detector network to dynamically update and manage travel paths, intersections, and energy usage based on real-time positional and elevation data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If pre-defined routes and static virtual models are used in fleet management systems, then system complexity is reduced and ease of operation is improved, but the ability to accurately detect and maintain changing travel routes deteriorates

Engineering Contradiction:
Improveease of operationVSAvoiddetection accuracy of changing routes
Core Design Contradiction:
Ease of operationVSAdaptability or versatility

Solution Approach 1:

The system transitions from static pre-defined routes to dynamic route detection by continuously collecting telemetry data from multiple machines and identifying commonly traveled paths. The virtual model automatically updates to reflect actual changing routes in the work site environment, resolving the contradiction between operational simplicity and route detection accuracy.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The fleet management system performs self-updating by automatically detecting roads and routes through telemetry data analysis without requiring manual surveying or model updates. The system uses machine learning algorithms to identify patterns in machine trajectories and autonomously maintains the virtual work site model, improving both ease of operation and adaptability.

Inventive Principle:
Principle #25Self-service

2Measurement precision

If manual surveying and geolocated topographic maps are used to create virtual models, then measurement precision is maintained, but productivity deteriorates due to increased effort and time required

Engineering Contradiction:
Improvevirtual model accuracyVSAvoidmodel creation and maintenance efficiency
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The system replaces manual surveying mechanics with electronic telemetry data collection and automated processing. GPS receivers and telematics devices on machines continuously provide position data that is processed by controllers to automatically update the virtual model, eliminating manual fieldwork while maintaining precision through multiple data sources and validation algorithms.

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

Solution Approach 2:

The virtual model creation and maintenance becomes a continuous automated process rather than periodic manual updates. Telemetry data is collected continuously as machines operate, and the system continuously processes this data to maintain an up-to-date virtual representation of the work site, significantly improving productivity while preserving measurement accuracy through ongoing validation.

Inventive Principle:
Principle #20Continuity of useful action

3Device complexity

If nodes are used to monitor machine location and travel in environments with changing routes, then device complexity is reduced, but reliability deteriorates when travel routes and operational locations change

Engineering Contradiction:
Improvemonitoring system complexityVSAvoidroute detection reliability
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The system replaces static node-based monitoring with dynamic route detection that adapts to changing operational conditions. Instead of relying on pre-positioned nodes that become obsolete when routes change, the system continuously analyzes telemetry data to identify actual machine paths and updates the virtual model accordingly, maintaining reliability without increasing complexity.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system implements feedback loops where telemetry data from machine trajectories is continuously fed back to the route detection algorithm. This feedback mechanism allows the system to learn from actual machine behavior, identify changing patterns, and automatically adjust the virtual model to reflect current operational routes, ensuring reliability in dynamic environments.

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS20250321106A1Systems and methods for the detection of roads
Publication Date: 2025.10.16 CATERPILLAR INC
  • US20250321106A1 patent drawing
  • US20250321106A1 patent drawing
  • US20250321106A1 patent drawing

AI summary

A system and method for generating or updating roads in a travel network of a work site. Telemetry data received from machines operating at the work site is received by a road determination module. The road determination module analyzes the telemetry data to find similar trajectories. The similar trajectories are identified as a road. The route determination module thereafter uses the roads generated or updated to manage the machines at the work site.