Key Note Speaker
Mel W. Torrie, Chief Executive Officer, Autonomous Solutions, Inc.
Presentation Topic: Keys to Success in Fielding Robotics
Summary: Autonomous Solutions, Inc. (ASI) has robot vehicles operating around the world in markets like mining, agriculture, automotive, security, and industrial automation and there have been many lessons learned. For 15 years, ASI has been a leader in robotics and vehicle automation and Mel Torrie will share ASI’s experiences in these endeavors with an emphasis on work in the agriculture industry.
Brief Biography: Mel W. Torrie, M.S., Chief Executive Officer, Autonomous Solutions, Inc. has a master’s degree in Electrical Engineering from Utah State University and is the founder and CEO of Autonomous Solutions, Inc. (ASI) in Logan Utah. ASI develops and sells systems of robots in Mining, Military, Agriculture, Material Handling, Automotive Proving Grounds and Industrial Cleaning. Prior to founding ASI 15 years ago, Mr. Torrie worked at Utah State University where he managed two NASA Space Shuttle Payloads. Mel is a sought after speaker in the robotics community and has delivered addresses at conference events such as the Precision Farming Expo, RoboBusiness, The Canadian Institute of Mining, Optimizing Mining Operations, Robotics Alley, and RISE Hong Kong.
Hugo Emmanuel, Regional Director, Irstea – Clermont-Ferrand Centre
Presentation Topic: How new technologies can help agriculture to face its major challenges
Brief Biography: After completing his studies at the Ecole nationale supérieure agronomique (National Agronomy Institute) in Rennes, France, Emmanuel Hugo joined Irstea (Cemagref at the time) in 1982 as a specialist in agricultural machinery and associated economics. As a result, he has held different positions and worked in particular on the theme of “evaluating agricultural machinery” in Antony and Montpellier centres. In 2007, he became leader of the research unit “Technologies and information systems for Irstea’s agricultural systems” at Irstea’s Clermont-Ferrand centre. He took over a position of regional director of this centre on 1 April 2012.
Jay Katupitiya, Associate Professor, UNSW Mechanical and Manufacturing Engineering, UNSW AUSTRALIA
Presentation Topic: Technologies to Achieve Precision, Scalability and Safety of Future Autonomous Agricultural Machines.
Summary: Although not a common occurrence, autonomous agricultural machines are slowly finding their way into the agricultural fields. In particular, unlike their urban counterparts, the autonomous machines used for broad acre farming operate in relatively safer and privately owned land free of clutter. The terrain on which they operate are most often machine friendly. In such situations, it is not only possible to build highly precise and reliable autonomous agricultural machines that would operate round the clock but also further research can be conducted to investigate innovative machine configurations that ensure factory level precision at high operating speeds with highly flexible scalability. The speaker will initially present the precision autonomous seeding systems he has developed, the commercial autonomous weeding systems that has implemented his algorithms and the ongoing development of autonomous bulldozers for Komatsu. Then the presentation will discuss potential future research directions to achieve scalability and precision and will highlight their research challenges.
In the case of broad acre farming, the machines used are gigantic and are very powerful. The autonomous machines of such proportions have the potential to cause major destruction if their guidance systems fail resulting in collision. Given that the technologies that are developed for autonomous agricultural machinery are directly applicable to mining machines which are even more powerful, attention to safe operation is of paramount importance. Hence there is a need to have procedures in place to verify and certify the safe operation of these machines. This presentation will highlight the two most common causes of such dangerous situations, one being the possibility of less than desired safety standards implemented on the autonomous agricultural machines by their manufacturers, and the other being the interference with the operation of the driverless machines by external parties which usually occur over the network. It is only natural that we have a tendency to maintain contact with the autonomous machine, mostly for the fear of it malfunctioning, hoping to bring the situation under control from afar, and in some other circumstances to have access to real-time data for various purposes, for example, for the visualization of the machine’s whereabouts on a map. This presentation will discuss in detail the partitioning of the information flow in a logical sense and will propose approaches to mitigate the potential undesirable behaviour. The presentation will also propose steps that can be taken to ensure the implementation of adequate reliability and safety standards by the manufacturers so that a cultural shift in the minds of general public can be developed by realizing the agricultural machines as trusted autonomous systems.
Brief Biography: Jay Katupitiya is an Associate Professor and Head of Mechatronic Engineering in the School of Mechanical and Manufacturing Engineering as well as affiliated with the Research Centre for Integrated Transport Innovation (rCITI), University of New South Wales.
Minzan Li, Professor, Research Center for Precision Agriculture, China Agricultural University
Presentation Topic: Recent Progresses in Precision Agriculture in China
Brief Biography: Professor Li received his Ph.D. degrees in 2000 in agricultural engineering from Tokyo University of Agriculture and Technology, Japan. He is currently the lead of the Research Center for Precision Agriculture at China Agricultural University.
Scott A. Shearer, Professor and Chair, Food, Agricultural and Biological Engineering Department, The Ohio State University
Presentation Topic: From Manned to Full Autonomy; Factors Influencing the Development and Adoption of Automation for Agricultural Field Machinery
Summary: The development of agricultural field machinery has been heavily influenced by technological developments in other sectors. For example, without defense-related concerns it is doubtful that the civilian sector alone would have justified deployment of space-based radio navigation or Global Navigation Satellite Systems (GNSS). Similarly, the transportation sector has contributed to the adoption and use of embedded controls, or controller area networks (CAN), in off-road equipment. The goal of this presentation is to examine a number of factors relating to the removal of man, as a control element, from field production systems. Many forces external to the industry will shape the evolution and adoption of automation in agriculture. Factors to be addressed in this presentation include: equipment physical limitations (transport); machine life versus technical obsolescence; soil quality and health; sensor densification; IoT; evolving automotive technologies; full versus supervised autonomy; product liability; broadband internet access in rural areas; off-highway emissions regulations; evolving machinery business models; and workforce development.
Brief Biography: Dr. Scott A. Shearer received formal training in agricultural engineering from The Ohio State University and was awarded B.S., M.S. and Ph.D. degrees in 1981, 1983 and 1986. Currently, he serves as Professor and Chair of the Food, Agricultural and Biological Engineering at The Ohio State University. Prior to OSU he was on a faculty member and then Chair of Biosystems and Agricultural Engineering at the University of Kentucky. During his 28-years in academia, his research efforts have focused on spectral and spatial image processing for the extraction of features for classifying agricultural settings; and controls and methodologies for metering and distribution of inputs (e.g., seed, fertilizer and chemicals) in grain crop production systems. His current research activities include autonomous multi-vehicle field production systems and unmanned aerial systems for remote sensing. He has lead or cooperated in research supported by more than $8M in grants; authored or coauthored more than 170+ refereed journal articles, conference proceedings, and technical publications; and has made numerous invited presentations at extension meetings, workshops, farmer forums, and outreach events.
Invited Talks for Technical Sessions:
Viacheslav Adamchuk, McGill University
Title: Low-Cost Smart Tractor Control Options
Summary: GNSS-based auto-guidance of agricultural tractors has become the most popular precision agriculture practice in North America. Simplicity of operation and immediate benefits have led to its rapid adoption. However, this is not the case with variable rate technologies. In many instances, an affordable level of GNSS positioning does not allow farmers to achieve the guidance accuracy required for specific field operations, e.g. organic row crop cultivation. This presentation provides an overview of two automated control techniques addressing the needs of small and medium size growers, who attempt to implement precision agricultural practices with limited budgets. Thus, the proposed smart tractor concept relates to the potential of employing existing tractor control hardware to pursue variable rate applications of agricultural inputs in response to supervised input variables defined by the operator. The so-called “zones on-the-go” system was developed to enable three different types of input (an internal sensor, an external sensor, and a prescription map) and three different control options (stage control, one-input proportional control, and two-inputs proportional control). All signal communication was conducted using ISOBUS. Even though several different tractor actuators could be operated using this system (e.g., PTO, hydraulic flow, 3-point hitch height), preliminary trials were performed on travel speed control. This control approach allowed the system to vary agricultural inputs using machines equipped for a constant rate of material discharge. The second technique involved the use of low-cost digital cameras to achieve accurate machinery guidance. An adds-on computer vision sensing system has been successfully implemented for in-row cultivator guidance with less than 5 cm error margin. In addition, a quick-attach steering column stepper motor assembly was developed to enable tractor’s guidance with 10 cm error margin when operating at speeds up to 15 km/h. Both techniques present interesting adaptive control challenges where the key parameters of the control algorithms are adjusted by the operator in real-time.