Zapisane w:
| 1. autor: | |
|---|---|
| Format: | Recurso digital |
| Język: | |
| Wydane: |
Zenodo
2026
|
| Hasła przedmiotowe: | |
| Dostęp online: | https://doi.org/10.5281/zenodo.19603700 |
| Etykiety: |
Dodaj etykietę
Nie ma etykietki, Dołącz pierwszą etykiete!
|
Spis treści:
- <p>The two uploaded documents together describe a complete forest-robotics solution built around a quadruped robot. The first document explains how the robot can be deployed in forests for patrol, monitoring, navigation, and long-duration autonomous missions. The second document explains how the data collected by that robot is processed into measurable forestry outputs such as tree detection, DBH estimation, GPS geolocation, and cross-referencing with drone data. In simple terms, the first document is the <strong>field deployment and operations handbook</strong>, while the second is the <strong>technical data-processing and inventory workflow</strong>.</p> <p><strong>Document 1: Quadruped Forest Patrol Handbook</strong><br>This document is a technical handbook for using the Boston Dynamics Spot quadruped robot in <strong>forest patrol and environmental monitoring</strong>. It focuses on how to deploy the robot in difficult forest terrain using a combination of stereo cameras, EAP2 LiDAR, GPS/GNSS, fiducial markers, GraphNav, and Autowalk. It explains the robot platform, navigation architecture, terrain handling, patrol mission design, power and charging strategy, communications options, risk mitigation, maintenance, and operational recommendations. It also includes a state-of-the-art review and appendices with stakeholder Q&A and lessons learned from the Kavala pilot.</p> <p>This handbook is useful for <strong>forest rangers, environmental agencies, conservation organizations, research institutions, robotic field operators, safety managers, and engineers</strong> planning or running autonomous forest missions. It is especially useful for organizations that want to know whether a quadruped robot is practical for patrol, inspection, and monitoring in real forest conditions, and what infrastructure is needed to make it reliable. The document itself says it is written for mixed audiences including field operators, safety leads, forestry stakeholders, and engineers.</p> <p> </p> <p><strong>Document 2: Ground-Based Forest Inventory Workflow</strong><br>This document is a technical workflow report for a <strong>ground-based robotic forest inventory system</strong>. Its purpose is to explain how the robot’s camera, LiDAR, GPS, and telemetry data are processed into forestry information. The workflow covers synchronized image–LiDAR–GPS capture, LiDAR–camera fusion, YOLO-based detection of trees and trunks, DBH estimation from point clouds, GPS geolocation of detections, and matching ground-measured trunks with drone-detected crowns. The document is organized around four software modules that together create an end-to-end processing pipeline.</p> <p>This document is useful for <strong>AI engineers, computer-vision researchers, LiDAR specialists, GIS teams, forestry inventory analysts, drone-survey teams, database engineers, and research organizations</strong> that want to process robot-collected forest data into actionable inventory records. It is less about field operations and more about the <strong>software and analytics pipeline</strong> that turns raw data into measurements and maps.</p> <p> </p> <p>Various videos from patrol demonstration and also Q&A videos , quadruped deployment lessons are included (best way to view the videos is to download them and use VLC video viewer.</p> <p> </p>