Skip to content

General Purpose Datasets with user specified sensor(s)

Contact Service Provider

Summary

Datasets are acquired in the agriculture field/forest with user specific sensors with adequate mobile robot (ground, aerial)

S00211
Version: 1.7

  • arable farming
  • horticulture
  • tree crops
  • viticulture
  • food processing
  • desk assessment
  • collection of test data
  • provision of datasets
  • data augmentation
  • data analysis
  • physical system
  • data
  • design/documentation
  • other (including: nothing)
  • Location: in France
  • Offered by: INRIA

E-Mail

Description

General purpose datasets target two main goals, i) Mobility Algorithms Evaluation, ii) General purpose AI applications development and evaluation. Mobility Algorithms concerns the classical robotics functionalities of mapping, localization, slam, and navigation. Whereas, general purpose AI application concerns the development of algorithms for weed detection, health monitoring, growth & maturity monitoring, and to feed decision support systems (DSS) for arable farming, horticulture, food processing, trees and forestry. Monitoring the environment poses a significant hurdle in devising algorithms and AI solutions for the applications in the agricultural robotics, as it is possible to have change in the environment of agriculture fields due to different seasons and weather conditions. The solution is to obtain consistent and periodic data to survey the dynamic changes in the agricultural landscape. This data is crucial in crafting efficient algorithms and AI solutions. It involves the systematic approach to collect the real-time data from agricultural fields using either a ground robot or an aerial robot equipped with an array of sensors, including Camera, LiDAR, IMU, and RTK-GPS along with the user specified sensor(s). The synchronization of these sensors through timestamps produces meticulously annotated data. This dataset can be leveraged for the development of techniques tailored to each sensor individually or harnessed for the creation of multisensor algorithms.