- Venue: WSCC 609
- Workshop was a big success. Slides and posters are posted below.
- Program is now online.
- Extended abstract submission deadline extended to April 7, 2015.
- Call for Papers is online. Deadline: March 30, 2015.
Agriculture faces a number of unique social, economic and environmental challenges. These include increasing globalisation and international competitiveness, climate variability, food security, shortages in labour, urban pressure on farmland, invasive species, and growing expectations on sustainability and land stewardship. Furthermore, the global population is expected to reach 9 billion by 2050, requiring agricultural production to double in order to meet food demands. Farm enterprises require new and innovative technology to address these challenges. Robotics is one of these technologies that promises to provide a solution.
This full-day workshop at ICRA '15 brings together researchers and industry working on novel approaches for long term operation across changing agricultural environments, including broad-acre crops, orchard crops, nurseries and greenhouses, and horticulture. The goal of the workshop is to discuss techniques that increase robustness and reliability for overall improvement of complete autonomy in the farm. This includes addressing the technology gap between the needs of the agriculture industry and existing robotic solutions and what the role of robotics researchers is in bridging this gap. Besides new concepts in perception for changes in appearance and geometry of the environment, new learning and adaptive approaches to novel environments (due to seasonal and weather changes or to adaptation for operation in a completely different crop) are also of special interest to the workshop.
This workshop is supported by the IEEE RAS TC on Agricultural Robotics and Automation.
- May 15th: Deadline to submit final versions of the 4-page extended abstract of accepted posters
- February 1st: Call for submission opens
March 30thApril 7th: Submission deadline
- April 15th: Notification of acceptance
- May 30th: Workshop at ICRA, Washington State Convention Center, Seattle WA, USA
Topics of Interest
Topics of interest to this workshop include, but are not necessarily limited to:
- Novel perception for agricultural robots including passive and active methods
- Approaches to low-cost sensing for day/night continuous operation
- Long-term autonomy and navigation in unstructured environments
- Perception for appearance and geometric change (possibly caused by seasonal and weather changes, or deployment in a new crop) for long-term autonomy
- Aerial and ground robotic platforms for soil/crop monitoring, prediction, and decision making
- Adaptive sampling and informative data collection
- Manipulators and platforms for soil preparation, seeding, crop protection, and harvesting
- Multi-robot systems
- Deployment strategies and user interfaces for end-users
- Standardized benchmarks and long-term datasets in changing agricultural environments
We explicitly encourage the submission of papers describing work in progress, preliminary results, novel concepts, and industry experiences. Contributions are expected from researchers working in different fields of robotics, computer vision, intelligent transportation systems, autonomous vehicles, manipulation, control, path planning, computer human interaction, machine learning, as well as end user and Original Equipment Manufacturer experiences, and technology integration in the agricultural workplace.
The workshop will consist of five invited talks and two poster sessions. The poster sessions consist of five-minute spotlight talks by the presenters, followed by an half-hour interactive session (coinciding with the morning and afternoon coffee breaks).Session 1
08:30 - 08:45 Welcome & introduction
08:45 - 09:30 Invited talk 1 – Tristan Perez (Queensland University of Technology) [slides]
09:30 - 10:00 Paper spotlight 1 (5 presentations, 5 min each)
10:00 - 10:30 Interactive session 1 + coffee
- Luca Carlone et al., Georgia Tech, “Towards 4D Crop Analysis in Precision Agriculture: Estimating Plant Height and Crown Radius over Time via Expectation-Maximization”
- Tristan Perez et al., Queensland University of Technology, “A Bayesian Framework for the Assessment of Vision-based Weed and Fruit Detection and Classification Algorithms” [slides]
- Chao Qu et al., University of Pennsylvania, “A Versatile Imaging System for Automated Monitoring enabling Precision Agriculture”
- Inkyu Sa et al., Queensland University of Technology, “On Visual Detection of Highly-occluded Objects for Harvesting Automation in Horticulture” [poster] [slides]
- Ce Yang et al., University of Minnesota, “Blueberry Maturity Stage Detection Based on Spectral-Spatial detection of Hyperspectral Image Using Selection Bands” [slides]
10:30 - 11:15 Invited talk 2 – Sanjiv Singh (Carnegie Mellon University)
11:15 - 12:00 Invited talk 3 – Joseph Jones (Harvest Automation) "Harvey on the Farm: Case study of an agricultural robot in the marketplace" [slides]
12:00 - 13:30 Lunch
13:30 - 14:15 Invited talk 4 – Wonsuk Lee (University of Florida) "Current status and challenges of sensing systems for precision agriculture in Florida"
14:15 - 15:00 Paper spotlight 2 (6 presentations, 5 min each)
15:00 - 15:30 Interactive session 2 + coffee
- David Anthony et al., University of Nebraska–Lincoln, “Towards GPS Free Localization in Row Crops” [slides]
- Joe Davidson et al., Washington State University, “Human-Machine Collaboration for the Robotic Harvesting of Fresh Market Apples”
- Chris Lehnert et al., Queensland University of Technology, “Optimisation-based Design of a Manipulator for Harvesting Capsicum” [slides]
- Scarlett Liu et al., The University of New South Wales, “A Lightweight Method for Grape Berry Counting based on Automated 3D Bunch Reconstruction from a Single Image” [slides]
- James Underwood et al., University of Sydney, “Real-time target detection and steerable spray for vegetable crops” [slides]
- Yawei Zhang et al., Oregon State University, “Intelligent In-Orchard Bin-Managing System for Tree Fruit Production” [poster] [slides]
15:30 - 16:15 Invited talk 5 – Stewart Moorehead (John Deere) “Autonomy in Agriculture – An Industry Perspective” [slides]
16:15 - 16:45 Workshop discussion
Sanjiv Singh. Research Professor, Robotics Institute; Courtesy Appointment, Mechanical Engineering
Dr. Singh is a Research Professor at the Robotics Institute with a courtesy appointment in Mechanical Engineering. He is the founding editor of the Journal of Field Robotics.
Prof. Singh’s research relates to the operation of robots in natural and in some cases, extreme environments. His recent work has two main themes: perception in natural and dynamic environments and multi-agent coordination. Currently he leads efforts in collision avoidance for air vehicles (near earth and between aircraft) and ground vehicles using novel sensors that can look through obscurants. Another research area seeks to provide accurate tracking and situational awareness in dynamic environments, such as those encountered in search and rescue, using radio signals to compute location. This research seeks to assist emergency response personnel in their operations. Currently, Prof. Singh leads a four-year, multi-institutional effort to investigate the use of automation in specialty agriculture. This effort is focused on obtaining high-resolution spatio-temporal data about high value crops in order to increase yield and minimize disease and infestations. The second theme is developing a multirobot architecture to perform complex tasks such as assembly that cannot be performed by a single machine. In this architecture, tasks are distributed between robots and human operators such that the strengths of each can be exploited and control can ‘slide’ effectively between humans and robots. Pursuant to this theme, Prof. Singh has developed a number of mobile manipulators to deal with dexterous, perception based manipulation tasks.
Joe Jones, Founder & CTO
Joe invents practical robots. A robotics visionary with 24 years in the field, Joe was the first employee of iRobot, where he and Harvest cofounder Paul Sandin invented Roomba, the floor-cleaning blockbuster, which has sold over 5 million units. He spent 9 years in robotic research at the MIT AI Lab, authored three books on robotics, and holds 15 patents.
Tristan Perez completed his Electronics Engineering degree (a 6-year program) at the National University of Rosario in Argentina in 1999 and his PhD in the area of control Engineering at the University of Newcastle, Australia, in 2003. In 2004, he moved to the UK and took a position as a Research Fellow at the Mechatronics Centre of the University of Wales, Newport, where he worked on fault diagnosis and control of propulsion systems for underwater vehicles – a collaboration with Seaeye Marine (SAAB). In July 2004, Tristan moved to Norway, where he spent three and a half years at the Norwegian University of Science and Technology (NTNU). He was a Senior Research Fellow at the CoE for Ships of Ocean Structures (CeSOS). At CeSOS, he conducted research on mathematical modelling of ship dynamics for manoeuvring in a seaway, ship roll stabilisation, and control allocation for ship dynamic positioning. Tristan was also a lecturer in marine control systems. In 2007, he moved back to Australia to the ARC-CoE for Complex Dynamic Systems and Control (CDSC) at the University of Newcastle (UoN). At CDSC, he developed a research program on dynamics and motion control of marine and aerospace vehicles. Within this program, he led 11 projects with 5 industry partners from Australia, New Zealand, Spain, and Norway. In 2010, Tristan was appointed A/Prof of Mechatronics at the University of Newcastle (UoN), where he undertook the re-development of the BEng Mechatronics program and developed the laboratory for autonomous systems and robotics.Also from 2009 to 2012, Tristan was appointed Adjoint A/Prof of Ship Dynamics at NTNU (Norway). In April 2014, Tristan was appointed Professor of Robotics at the School of Electrical Eng. & Computer Science at QUT where he is working on a projects related to Agricultural robotics; Bio-inspired guidance and motion control of unmanned aircraft (collaboration with UQ and Boeing); Energy-based modelling and motion control of field robots (collaboration with UoN and DSTO)
Wonsuk "Daniel" Lee
Agricultural and Biological Engineering Department, University of Florida
- Ph.D., Biological and Agricultural Engineering, University of California, Davis, 1998. Dissertation: Robotic Weed Control System for Tomatoes.
- Master of Science., Agricultural Engineering, Seoul National University, Suwon, Korea, February 1988. Thesis: A single layer drying equation for red pepper.
- Bachelor of Science., Agricultural Engineering, Seoul National University, Suwon, Korea, February 1986.
- Associate Professor, University of Florida, 7/2006 - present.
- Assistant Professor, University of Florida, 5/2000 - 6/2006.
RESEARCH INTERESTS: Precision Agriculture; Instrumentation and Control; Machine vision and image processing in real-time for robotic weed control; Field plant identification and pattern recognition for farm automation; Development of new technologies for weed control to reduce or remove agricultural chemical application; Computer interfacing and signal processing; Fuzzy and Neural Network; Near-infrared application for agricultural products.
Precision Agriculture Laboratory: Soil nutrient sensing; Citrus nitrogen sensing; Phosphorus sensing in the Lake Okeechobee drainage basins; Citrus yield mapping using machine vision; Green citrus detection; Silage yield mapping; Precision farming adoption by Florida citrus producers
Deere & Company, Manager, Robotic Systems
Dr. Moorehead is Manager for Robotic Systems at John Deere Technology. Champaign, IL. In this role, he is responsible for leading the enterprise efforts in robotics technology development and coordinating activities with the business units. Stewart joined John Deere in 2004 to work on the Autonomous Stadium Mower project. Since then, he has been involved in several robotics projects including the R-Gator and the Autonomous Orchard Tractor. His work has focused on obstacle detection and improving the robustness of robotics technology to function in unforgiving conditions found on farms and construction sites. Stewart has been on the Board of Directors of the AUVSI since 2013 and is a member of the AUVSI Agricultural sub-committee. He is also an active member of the IEEE Robotics and Automation Society’s Agricultural Robotics and Automation Technical Committee. Stewart has a PhD in Robotics from Carnegie Mellon University and Bachelor and Masters degrees in Electrical Engineering from the University of Waterloo.