gmdh application for autonomous mobile robot’s control system construction
DESCRIPTION
GMDH Application for autonomous mobile robot’s control system construction. A . V . Tyryshkin , A . A . Andrakhanov , A . A . Orlov Tomsk State University of Control Systems and Radioelectronics E-mail: [email protected]. Classification of existing autonomous robots. - PowerPoint PPT PresentationTRANSCRIPT
![Page 1: GMDH Application for autonomous mobile robot’s control system construction](https://reader033.vdocument.in/reader033/viewer/2022061600/5681479f550346895db4d8b9/html5/thumbnails/1.jpg)
GMDH Application for autonomous mobile robot’s control system construction
A.V. Tyryshkin, A.A. Andrakhanov, A.A. Orlov
Tomsk State University of Control Systems and Radioelectronics
E-mail:E-mail: [email protected]@mail.ru
![Page 2: GMDH Application for autonomous mobile robot’s control system construction](https://reader033.vdocument.in/reader033/viewer/2022061600/5681479f550346895db4d8b9/html5/thumbnails/2.jpg)
Classification of existing autonomous robots
![Page 3: GMDH Application for autonomous mobile robot’s control system construction](https://reader033.vdocument.in/reader033/viewer/2022061600/5681479f550346895db4d8b9/html5/thumbnails/3.jpg)
Nearest analog – agricultural AMR “Lukas”
![Page 4: GMDH Application for autonomous mobile robot’s control system construction](https://reader033.vdocument.in/reader033/viewer/2022061600/5681479f550346895db4d8b9/html5/thumbnails/4.jpg)
![Page 5: GMDH Application for autonomous mobile robot’s control system construction](https://reader033.vdocument.in/reader033/viewer/2022061600/5681479f550346895db4d8b9/html5/thumbnails/5.jpg)
Basic works on GMDH application to AMR control
C.L. Philip Chen, A.D. McAulay Robot Kinematics Learning Computations Using Polynomial Neural Networks, 1991;
C.L. Philip Chen, A.D. McAulayRobot Kinematics Computations Using GMDH Learning Strategy, 1991;
F. Ahmed, C.L. Philip ChenAn Efficient Obstacle Avoidance Scheme in Mobile Robot Path Planning using Polynomial Neural Networks, 1993;
C.L. Philip Chen, F. AhmedPolynomial Neural Networks Based Mobile Robot Path Planning, 1993;
A.F. Foka, P.E. TrahaniasPredictive Autonomous Robot Navigation, 2002;
T. Kobayashi, K. Onji, J. Imae, G. Zhai Nonliner Control for Autonomous Underwater Vehicles Using Group Method of Data Handling, 2007;
![Page 6: GMDH Application for autonomous mobile robot’s control system construction](https://reader033.vdocument.in/reader033/viewer/2022061600/5681479f550346895db4d8b9/html5/thumbnails/6.jpg)
Part I Inductive approach to
construction of AMR control systems
![Page 7: GMDH Application for autonomous mobile robot’s control system construction](https://reader033.vdocument.in/reader033/viewer/2022061600/5681479f550346895db4d8b9/html5/thumbnails/7.jpg)
Problems of AMR design
Navigation Obstacle Recognition Autonomous Energy Supply Optimal Final Elements Control Technical State Diagnostics Objectives Execution Knowledge Gathering and Adaptation
![Page 8: GMDH Application for autonomous mobile robot’s control system construction](https://reader033.vdocument.in/reader033/viewer/2022061600/5681479f550346895db4d8b9/html5/thumbnails/8.jpg)
Generalized structure of AMR
![Page 9: GMDH Application for autonomous mobile robot’s control system construction](https://reader033.vdocument.in/reader033/viewer/2022061600/5681479f550346895db4d8b9/html5/thumbnails/9.jpg)
Objective aspects of AMR control system construction
Utility
Realizability
Appropriateness
Classification
Taking into account Internal system parameters
Forecasting
![Page 10: GMDH Application for autonomous mobile robot’s control system construction](https://reader033.vdocument.in/reader033/viewer/2022061600/5681479f550346895db4d8b9/html5/thumbnails/10.jpg)
Features of AMR obstacle recognition
Lack of objects’ a priori information Objects to recognize are complex ill-conditioned
systems with fuzzy characteristics Objects are characterized by high amount of
difficultly- measurable parameters
It is necessary to take into account internal systems parameters for objects’ classification according to “obstacle/not obstacle” property, i.e. it isn’t possible to find out is this object obstacle or not without regard for system state.
There is no necessity to perform full object identification, i.e. it isn’t necessary to answer a question “What object is this?”
![Page 11: GMDH Application for autonomous mobile robot’s control system construction](https://reader033.vdocument.in/reader033/viewer/2022061600/5681479f550346895db4d8b9/html5/thumbnails/11.jpg)
![Page 12: GMDH Application for autonomous mobile robot’s control system construction](https://reader033.vdocument.in/reader033/viewer/2022061600/5681479f550346895db4d8b9/html5/thumbnails/12.jpg)
Part IIAutonomous Cranberry
Harvester
![Page 13: GMDH Application for autonomous mobile robot’s control system construction](https://reader033.vdocument.in/reader033/viewer/2022061600/5681479f550346895db4d8b9/html5/thumbnails/13.jpg)
Expected Engineering-and-economical Performance
Nominal Average AMR speed:
Cranberry harvesting coverage:
Relative density of harvested cranberry:
Total weight of harvested cranberry per season:
Season income:
kgdaysdayh
hkg 1440003010480season
4608002.3144000 kgUSDkg$$ USD
hkg
mkg
hmPrel 4801.04800 22
hmmh
mSharvest2
48002.14000
hkm
nom 4
![Page 14: GMDH Application for autonomous mobile robot’s control system construction](https://reader033.vdocument.in/reader033/viewer/2022061600/5681479f550346895db4d8b9/html5/thumbnails/14.jpg)
Automated cranberry harvester
![Page 15: GMDH Application for autonomous mobile robot’s control system construction](https://reader033.vdocument.in/reader033/viewer/2022061600/5681479f550346895db4d8b9/html5/thumbnails/15.jpg)
Part IIISimulation Results
![Page 16: GMDH Application for autonomous mobile robot’s control system construction](https://reader033.vdocument.in/reader033/viewer/2022061600/5681479f550346895db4d8b9/html5/thumbnails/16.jpg)
Object Recognition Data Sample
Learning samples – 92; Training samples – 50.
Values’ Ranges:
Object Length L Є [0;20] м;
Object Width w Є [0;20] м;
Object Height h Є [0;20] м;
![Page 17: GMDH Application for autonomous mobile robot’s control system construction](https://reader033.vdocument.in/reader033/viewer/2022061600/5681479f550346895db4d8b9/html5/thumbnails/17.jpg)
Recognizing Modified Polynomial Neural Network
31
32
231
232
32
31
41
24
224
224
31
22
222
222
32
13
14
213
214
14
13
24
12
14
212
214
14
12
22
227514
227513
2212
3386.36605.00329.05832.12282.33182.1
5629.143526.0296.70054.03085.0290.6
7788.62438.00244.30060.02089.07575.2
631.208304.50856.73786.00954.7692.11
054.260944.79610.78830.1213.11805.14
0500.00003.00003.010500.2104863.99573.0
0700.00004.00033.010157.210569.47673.0
0855.00707.00062.00034.00071.07512.0
FFFFFFF
FLFLLFF
FLFLLFF
FFFFFFF
FFFFFFF
LtLttLF
hthtthF
whwhhwF
![Page 18: GMDH Application for autonomous mobile robot’s control system construction](https://reader033.vdocument.in/reader033/viewer/2022061600/5681479f550346895db4d8b9/html5/thumbnails/18.jpg)
Objective Functions’ Data Sample
Learning samples – 140; Training samples – 140.
Values’ Ranges:
Surface density of cranberry distribution ρcranberry Є [0;1] kg/m2;
Cranberry harvesting efficiency η Є [20;75] %;
Average AMR speed Vaverage Є [0;7] km/h;
Nominal average AMR speed Vnomaverage Є [2;4] km/h;
AMR engine fuel consumption per 100 km Pfuel Є [150;600] liters/100 km.
Values’ laws of variation:
![Page 19: GMDH Application for autonomous mobile robot’s control system construction](https://reader033.vdocument.in/reader033/viewer/2022061600/5681479f550346895db4d8b9/html5/thumbnails/19.jpg)
Objective Functions
22 012.086.11057.0, tVtVVtmF averageaveragecranberryaveragecranberrycranberry
tVVtmF averagecranberryaveragecranberrycranberry 223 693.077.1110684.6,
fuel
cranberryfuel
averagecranberryfuelcranberry PPVmmF
11257
14.3710874.4, 223
Function of maximal cranberry harvest in preset time:
Function of maximal cranberry harvest in minimal time:
Function of maximal cranberry harvest with minimal fuel consumption:
![Page 20: GMDH Application for autonomous mobile robot’s control system construction](https://reader033.vdocument.in/reader033/viewer/2022061600/5681479f550346895db4d8b9/html5/thumbnails/20.jpg)
Main Indices of Simulation Data
CRPercentage of
Errors
0.055 12%
F(mcranberry,Δt) F(mcranberry,t) F(mcranberry, mfuel)
CR BS CR BS CR BS
3.8e-4 9.8e-3 8.6e-3 0.9 1.8e-3 1.6
1) Obstacle recognition criterion values
2) Objective Functions criterion values
![Page 21: GMDH Application for autonomous mobile robot’s control system construction](https://reader033.vdocument.in/reader033/viewer/2022061600/5681479f550346895db4d8b9/html5/thumbnails/21.jpg)
“Man should grant a maximal freedom to the computing machinery. Like a horseman having lost a way leave it to a discretion of his horse...”
A.G. Ivakhnenko. “Long-term forecasting and complex system control”, Publ. “Технiка”, Kiev, 1975. – p. 8.
![Page 22: GMDH Application for autonomous mobile robot’s control system construction](https://reader033.vdocument.in/reader033/viewer/2022061600/5681479f550346895db4d8b9/html5/thumbnails/22.jpg)
![Page 23: GMDH Application for autonomous mobile robot’s control system construction](https://reader033.vdocument.in/reader033/viewer/2022061600/5681479f550346895db4d8b9/html5/thumbnails/23.jpg)
Нахождение разделяющих областей в пространстве параметров распознавания
Пространство параметровраспознавания
Область объектов-непрепятствий
Область условнопреодолимых препятствий
Область объектов-препятствий
`
L
h
![Page 24: GMDH Application for autonomous mobile robot’s control system construction](https://reader033.vdocument.in/reader033/viewer/2022061600/5681479f550346895db4d8b9/html5/thumbnails/24.jpg)
Современные состояние разработок в области АПК
![Page 25: GMDH Application for autonomous mobile robot’s control system construction](https://reader033.vdocument.in/reader033/viewer/2022061600/5681479f550346895db4d8b9/html5/thumbnails/25.jpg)
Итерационный алгоритм МГУА с разделением обучения