Autonomous and Human- Robot Collaborative Systems for Field Operations in Orchards, Greenhouses and Field Crops Avital Bechar Institute of Agricultural Engineering, ARO, Volcani Center, Israel 1
Overview Background The Agricultural Research Organization Agricultural productivity and production (robotics perspectives) Characteristics of the agricultural domain (robotics perspectives) Basic principles (AgRobots) ARL activity Conclusions AgRA TC Webinar, March 2 24 , 2015
Agricultural Research Organization AgRA TC Webinar, March 3 24 , 2015
Agricultural Research Organization • Founded in 1921. • 1000 people: including 200 research scientists and 220 graduate students. • 6 Institutes: Soil water and environmental sciences; Plant protection; Animal Sciences; Plant sciences; Food sciences; and, Agricultural Engineering. AgRA TC Webinar, March 4 24 , 2015
Institute of Agricultural Engineering The only research organization in Israel whose activities encompass a wide range of engineering and technological topics relating to all aspects of agriculture. About 60 people, including 14 research scientists AgRA TC Webinar, March 5 24 , 2015
Institute of Agricultural Engineering Two departments: Sensing, information, and mechanization engineering Production, growing and environmental engineering a AgRA TC Webinar, March 6 24 , 2015
Agricultural production Cultivation and production processes in agriculture. Affecting factors: crop characteristics and requirements, the geographical/geological environments, climatic conditions, market demands the farmer’s capabilities and means. Farm sizes increase and the number of farmers and agricultural workers decreases. Human labor intensive and labor cost of 25-40%. AgRA TC Webinar, March 7 24 , 2015
(http://www.thadw.us/agricultural-employment-since- 1870 / ) AgRA TC Webinar, March 8 24 , 2015
CV of different materials 0.55 0.5 0.45 0.4 CV= σ / µ 0.35 0.3 CV CV2 > CV1 0.25 0.2 0.15 0.1 0.05 0 Metal Metal Metal Metal Plastic Rubber Wood Flower Bolts screw nuts nails Discs parts parts parts cuttings AgRA TC Webinar, March 9 24 , 2015
AgRA TC Webinar, March 10 24 , 2015
AgRA TC Webinar, March 11 24 , 2015
Unstructured Environments • Unknown a-priori • Unpredictable • Dynamic AgRA TC Webinar, March 12 24 , 2015
Unstructured Environments The terrain, vegetation, landscape, visibility, illumination and other atmospheric conditions are not well defined; vary, have inherent uncertainty, and generate unpredictable and dynamic situations. AgRA TC Webinar, March 13 24 , 2015
Unstructured Objects Variable and non-uniform: size shape color texture location AgRA TC Webinar, March 14 24 , 2015
Industry Space Medical Agr. Under-water Military Env. + + - - Objects + + - - 15 AgRA TC Webinar, March 24 , 2015
Basic principles Main task: pruning, picking, harvesting, weeding... Supporting tasks: localization, detection, navigation… Mobility and steering Sensing Path planning and navigation Manipulators and end effectors Control Autonomy and human-robot collaboration AgRA TC Webinar, March 16 24 , 2015
Autonomy/Human-Robot collaboration Autonomous robot are lack the capability to respond to ill-defined, unknown, changing, and unpredicted events, such as occur in unstructured environments. Pareto principle: roughly 80% of a task is easy to adapt to robotics and automation and 20% is difficult (Stentz et al., 2002). AgRA TC Webinar, March 17 24 , 2015
Hybrid Human-Robot Systems Supporting Task 1 Supporting Supporting Main Task Task Task 4 2 Subsystem Subsystem Supporting 1 2 Task 3 AgRA TC Webinar, March 18 24 , 2015
AgRA TC Webinar, March 19 24 , 2015
Lab members (current) 2 PhD students (IE, CE-AgEng) 3 MSc students (ME, IE) Mechanical Engineer Electrical Engineer Postdoc Agronomist AgRA TC Webinar, March 20 24 , 2015
Projects (current) Autonomous greenhouse sprayer for specialty crops ( with BGU ). A human-robot collaborative system for deciduous tree selective pruning. a human-robot system for selective melon collection ( with Technion ). an autonomous system for monitoring of diseases in greenhouses ( with BGU ). Robotic sonar for yield estimation ( with TAU ). Characterization of Agricultural Tasks for the Design of a Minimalistic Robot ( with Technion ). AgRA TC Webinar, March 21 24 , 2015
Autonomous greenhouse sprayer Avital Bechar, Itamar Dar, Victor Bloch, Yael Edan, Roee Finkelshtein, Guy Lidor, Ron Berenstein AgRA TC Webinar, March 22 24 , 2015
The motivation AgRA TC Webinar, March 23 24 , 2015
Plot geometry 170 100 m 115 AgRA TC Webinar, March 24 24 , 2015
Sensing (Features) ∆ R R AgRA TC Webinar, March 25 24 , 2015
Features Feature Formula Feature Formula R Red h H/(H+S+V) G Green s S/(H+S+V) B Blue v V/(H+S+V) r R/(R+G+B) deltaH (H-S)+(S-V) g G/(R+G+B) deltaS (S-H)+(S-V) b B/(R+G+B) deltaV (V-S)+(V-H) deltaR (R-G)+(R-B) C1 R-G deltaG (G-R)+(G-B) C2 R-B deltaB (B-G)+(B-R) C3 G-B − − H Hue Real_ModHue − 2 R G B { 1 cos ( ) + + − − − 2 2 2 2 ( R G B RG RB GB ) S Saturation imag_ModHue V Value 26
Decision Tree - CART Breiman et al., 1984 For all features Find feature threshold value that maximizes the "splitting criterion“ Among all features Choose the one that maximizes the "splitting criterion“ AgRA TC Webinar, March 27 24 , 2015
Decision tree Total success 1 Level 2 Level 3 Level Movie TS TS TS Movie 1 0.834 0.834 0.886 Movie 2 0.943 0.941 0.940 Movie 3 0.617 0.834 0.848 Movie 4 0.818 0.874 0.889 Movie 5 0.922 0.927 0.920 Movie 6 0.892 0.899 0.899 Movie 7 0.932 0.925 0.930 Average 0.851 0.891 0.902 Number of nodes 1 3 7 AgRA TC Webinar, March 28 24 , 2015
Judges Vote (~ Majority rule) A customized CART variation, developed in this research A “Judge” is single level CART (root node only) Classification rule: Judges _ Vote Number _ of _ Judges Vote (M) 1 2 1 2 3 1 2 3 4 1 2 3 4 5 Judges (N) 2 2 3 3 3 4 4 4 4 5 5 5 5 5 AgRA TC Webinar, March 29 24 , 2015
Test set – "Judges Vote" Variation 2/2 2/3 3/4 3/5 4/5 2 Level (3 features) Average 0.903 0.914 0.915 0.905 0.890 0.920 std 0.041 0.021 0.016 0.020 0.022 0.044 AgRA TC Webinar, March 30 24 , 2015
Algorithm Evaluation Platform lifeCam NX-6000 Servo SC-1256T 45 Lenovo R400 PWM DAT A AX3500 - Dual 60A 180 ⁰ 123 CMP-03 Compass Arduno Encoder Optical Motor DL-30 AgRA TC Webinar, March E5 31 24 , 2015
AgRA TC Webinar, March 32 24 , 2015
TXT1 Ein Yahav 261109 1st exp-fast.wmv AgRA TC Webinar, March 33 24 , 2015
Modification of a commercial sprayer An electric motor was installed on the steering wheel controlled by a Roboteq controller. Installation of encoders on the steering pivot/axle and the front wheels. PID control system. Control system inputs: platform steering angle; desired direction from the adaptive algorithm and bearing. Pure pursuit, carrot point 2m AgRA TC Webinar, March 34 24 , 2015
The ‘autonomous unit’ Installed on the platform Connected to sensors and actuators AgRA TC Webinar, March 35 24 , 2015
Commercial Sprayer II AgRA TC Webinar, March 36 24 , 2015
AgRA TC Webinar, March 37 24 , 2015
A H-R collaborative system for selective pruning Avital Bechar, Victor Bloch, Roee Finkelshtain, Sivan Levi AgRA TC Webinar, March 38 24 , 2015
Objective Develop a human-robot integrated system for tree pruning and shaping Design of a cutting tool Develop a modelling technique Development of human robot interface and methodology AgRA TC Webinar, March 39 24 , 2015
Cutting tool alternatives Chain saw Pruning shears Laser Water jet Disc saw Jigsaw AgRA TC Webinar, March 40 24 , 2015
Cutting tool design The cutting tool must be adapted to: Tree dimensions, branch diameter and strength Robot carrying ability, precision, energy source Pruning technique: cutting angle, velocity Tree structure: branch angles, depth inside the crown, obstacle density, reaching ability Agronomical requirements: Cutting angle 45° Reduce risk of wounds Cut disinfection (burned by high cutting speed) AgRA TC Webinar, March 41 24 , 2015
Cutting tool selection & modification for a robotic arm Energy source, type and consumption Safety Weight Dimensions Precision and accuracy … AgRA TC Webinar, March 42 24 , 2015
High accuracy requirements Pruning shears: 3 directional dim. and 2 angular dim. Total required accuracy in 5D. Disk saw: 1 directional dim. and 2 angular dim. Total required accuracy in 3D AgRA TC Webinar, March 43 24 , 2015
Recommend
More recommend