Jos Balendonck from Wageningen University & Research and Ruud Barth from Saia Agrobotics, the Netherlands, presented a webinar on April 18, 2019 at 3:30pm/15:30 CET (UTC +2).
A video of the webinar is available at this link [207 MB].
Talk details.
Sweeper, a sweet pepper harvesting robot, operating in a greenhouse.
Title: Sweeper : A Sweet-pepper Harvesting Robot.
Abstract: In modern greenhouses there is a high demand to automate labor. The availability of a skilled workforce that accepts repetitive tasks in the harsh climate conditions of a greenhouse is decreasing rapidly. The current state of the art in automated harvesting of fruits and vegetables has remained remarkably stationary in the past decades. In the EU-FP7-project CROPS (www.crops-robots.eu) extensive research has been performed on agricultural robotics. One of the applications was a sweet pepper harvesting robot. In the H2020 project SWEEPER the technology developed in CROPS is used to introduce, test and validate a robotic harvesting solution for sweet pepper under real-world conditions. This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 644313.
Part I of this webinar presents the overall development, testing and validation of SWEEPER, a robot for harvesting sweet pepper fruit in greenhouses. The robotic system includes a 6 DOF industrial arm equipped with a specially designed end-effector, RGB-D camera, high end computer with GPU, PLCs, other electronic equipment, and a small container to store harvested fruit. All is mounted on a cart that autonomously drives on pipe rails and concrete floor in the end-user environment. The overall operation of the harvesting robot is described along with details of the algorithms for fruit detection and localization, grasp pose estimation, and motion control.
Part II of this webinar will go into depth of the vision principles for harvest robotics of this robot. A method for modelling 3D plants was used to generate large amounts of synthetic training data. This was then used to bootstrap Deep Learning networks to achieve state-of-the-art results on plant part segmentation.
Supplementary/Reading materials :
Project website: www.sweeper-robot.eu.
Barth, Ruud. Vision principles for harvest robotics : sowing artificial intelligence in agriculture, 2018, Wageningen University, PhD Thesis. link.
Biographical Information.
Jos Balendonck, MSc has an MSc degree in electronic engineering. He is a senior researcher on wireless and dielectric sensor technology, bio-monitoring and control applications in the area of robotics, climate and irrigation within a group of greenhouse horticulture. He coordinated two European projects on efficient irrigation management involving sensor technologies: FLOW-AID (FP6) and WATERMAN (FP4). Projects focused on the reduction of water and nutrient emissions from greenhouses related to the Water Framework Directive; sensor and model-based irrigation scheduling when using low-quality or saline water; soil humidity and salinity dielectric sensors (co-inventor of the WET-sensor). He further coordinated the European project on a sweet pepper robotic harvester (SWEEPER, H2020). He worked and is working on several (inter-) national projects devoted to plant monitoring and robotic harvesting (f.i. cucumber harvester).
dr. Ruud Barth has a Ph.D. degree (cum laude) in 2018 from Wageningen University on the topic Computer Vision Principles for Harvest Robotics, and a MSc degree (cum laude) in Artificial Intelligence from the Radboud University Nijmegen. He previously was a researcher and project leader at Wageningen Research where his studies focused on applied computer vision & robotics in the field of life sciences, including the latest deep-learning techniques. He currently is the CEO and founder of the spin-off company Saia Agrobotics that aims to bring the Sweeper robot technology to the market.