Understanding the requirements for a blind-spot monitoring system on tractors from the operator’s perspective

Canadian Biosystems Engineering

article

Xin Chen¹  Danny D. Mann¹ 

[1] Department of Biosystems Engineering, University of Manitoba

关键词: Blind spot;    Run-overs;    Tractor;    Testing;    Diagrams;    Camera;    Detection;    Agricultural machine;    Mirrors;    Sensors;    HAAT model;   

DOI  :  10.7451/CBE.2018.60.2.33

学科分类：农业科学（综合）

来源: Canadian Society of Agricultural Engineering

【 摘 要 】

Unintentional run-overs occur because the operator of a tractor isunable to physically see all around the machine. Therefore, thereis a need to devise an effective blind-spot monitoring system fortractors to prevent unintentional run-overs. The purpose of thestudy was to identify the locations of blind spots around two typesof tractors (i.e., with and without a front-end loader), with theultimate goal of conceptualizing a blind-spot monitoring systemcapable of eliminating all existing blind spots. Grids wereconstructed around all four sides of the tractors to determine thepresence of blind spots for drivers of varying sitting height (i.e.,5th, 50th and 95th percentile male for erect and slumped postures)at four horizontal planes representing people of varying staturewho might be in the vicinity of the tractor (i.e., standing male,standing female, standing child, kneeling adult). Generally, theproportion of markers not visible decreased as the sitting heightincreased. Differences between erect and slumped sitting postureswere not statistically different suggesting this variable could beignored in the assessment of blind spots around tractors. Theproportion of the markers not visible to the operator varied from 0to 34%, with higher values observed for the tractor with the frontend loader installed. Values were as high as 42% of the markersnot visible for the condition where a passenger was present in thepassenger/trainer seat. Use of the existing rear-view mirrorseliminated only a small fraction of the blind spot area around thetractors. Through trial and error, it was determined that five andeight cameras would be required to fully detect the entire blindspot area around the two tractors selected for this study. A blindspot monitoring system composed of five or eight cameras wouldcreate substantial additional monitoring burden for the tractoroperator and, therefore, is not a feasible solution. A hybrid blindspot monitoring system consisting of cameras and proximitysensors warrants further investigation.

【 授权许可】

CC BY