Ploughing Furrow Detection Based on 3D LiDAR Sensor Data

Reinhold Poscher, Riikka Soitinaho, Timo Oksanen

Research output: Contribution to journalConference articlepeer-review

Abstract

Even if today's powerful machines make ploughing easier and faster than before, it is still a demanding task for the operator. To reduce the driver's workload, improve safety, and increase performance, it is reasonable to automate the process. In this paper, we present a novel method for ploughing furrow edge detection based on data from a Velodyne VLP-16 3D Light Detection and Ranging (LiDAR) sensor. From this detection it is possible to derive a guidance line for navigation. Our method is based on maximum gradient edge detection on each individual line of the 16-layer 3D LiDAR data, and RANSAC for fitting a line to the detected edge points. The method is also capable of determining on which side of the furrow is the ploughed and the unploughed land. The evaluation of the method was conducted on a data set that was recorded during in-furrow ploughing, on the part of the data where the furrow edge is always visible on all lines of the LiDAR scan. Based on an initial evaluation the method can detect the ploughing furrow in 3D LiDAR point cloud data.

Original languageEnglish
Pages (from-to)54-59
Number of pages6
JournalIFAC Proceedings Volumes (IFAC-PapersOnline)
Volume55
Issue number32
DOIs
StatePublished - 2022
Externally publishedYes
Event7th IFAC Conference on Sensing, Control and Automation Technologies for Agriculture, AGRICONTROL 2022 - Munich, Germany
Duration: 14 Sep 202216 Sep 2022

Keywords

  • agricultural machinery
  • automatic guidance
  • guidance directrix
  • perception
  • sensing
  • tractors

Fingerprint

Dive into the research topics of 'Ploughing Furrow Detection Based on 3D LiDAR Sensor Data'. Together they form a unique fingerprint.

Cite this