Design of an interface for remotely supervised autonomous agricultural sprayers
Date
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
A study that aimed at designing a user interface for an autonomous agricultural sprayer was undertaken. It involved i) a survey of the farming community to gather their opinions about technological advancement of current agricultural machines and their expectations with respect to automated agricultural machines, ii) interviews with automated system designers and experts to understand how they intend/expect autonomous agricultural machines to be remotely supervised, iii) field and laboratory experiments to identify the necessary visual information for the remote supervisor to understand the operation of the automated machine, iv) ranking of machine and environment parameters based on their frequency of usage during machine monitoring, v) determining a suitable modality that will alert the supervisor of an issue requiring human attention and, vi) an evaluation of an automation interface. The survey of the farming community provided evidence that farmers and custom applicators are satisfied with the technological advancement of agricultural machines and are also willing to accept an automated sprayer when it becomes commercially available. The survey also provided an understanding of how farmers will prefer to interact with the automated machine. The field and laboratory experiments identified different regions of the machine and its environment that should be visually provided to the remote supervisor to enhance their understanding of the operation. A variety of remote supervision concepts were discovered during the interviews with designers of automated systems. These concepts were grouped into four categories based on the location of the human or remote station (within-the-field, close-to-the-field, farm office, and outside-the-farmland). Using the unranked paired analysis, the close-to-the-field remote supervision concept was considered the most viable concept. The study to identify the most suitable warning modalities revealed that the most suitable warning modality for the remote supervision concepts that had tractor sound in the background (i.e., within-the-field and close-to-the-field), was a combination of tactile and visual modalities (i.e., visual-tactile warning). An automation interface was designed for an agricultural sprayer using both the results from the requirement analysis and interface design guidelines. An evaluation of the interface revealed several strengths of the design as well as areas that needed further improvement.
Description
Keywords
Citation
Edet, U., E. Hawley and D. D. Mann. 2018a. Remote supervision of autonomous agricultural sprayers: the farmer’s perspective. Canadian Biosystems Engineering/Le génie des biosystèmes au Canada 60: 2.19-2.31. https://doi.org/10.7451/CBE.2018.60.2.19. Source: Canadian Society for Bioengineering.
Edet, U. and D.D. Mann. 2020. Visual information requirements for remotely supervised autonomous agricultural machines. Applied Sciences 10,2794. https:// doi:10.3390/app10082794. Source: MDPI Open Access Journal.