Slip estimation methods for proprioceptive terrain classification using tracked mobile robots

Ditebogo Masha; Michael Burke; Bhekisipho  Twala

doi:10.1109/RoboMech.2017.8261139

Slip estimation methods for proprioceptive terrain classification using tracked mobile robots

Ditebogo Masha, Michael Burke, Bhekisipho Twala

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Recent work has shown that proprioceptive measurements such as terrain slip can be used for terrain classification. This paper investigates the suitability of four simple slip estimation methods for differentiating between indoor and outdoor terrain surfaces, namely: rocks, grass, rubber and carpet. These slip estimates are calculated using experimental odometric data collected from a tracked autonomous ground vehicle and comprise of two instantaneous estimators and a temporal windowing approach. Results show that only the temporal windowing approach shows significant differences across the terrains investigated, indicating that instantaneous measurements are unsuited to terrain classification.

Original language	English
Title of host publication	2017 Pattern Recognition Association of South Africa and Robotics and Mechatronics
Place of Publication	Bloemfontein, South Africa
Publisher	Institute of Electrical and Electronics Engineers
Pages	150-152
Number of pages	3
ISBN (Electronic)	978-1-5386-2314-5
ISBN (Print)	978-1-5386-2315-2
DOIs	https://doi.org/10.1109/RoboMech.2017.8261139
Publication status	Published - 18 Jan 2018
Event	Symposium of the Pattern Recognition Association of South Africa (PRASA) and Conference of Robotics and Mechatronics (RobMech) - Bloemfontein, South Africa Duration: 29 Nov 2017 → 1 Dec 2017

Conference

Conference	Symposium of the Pattern Recognition Association of South Africa (PRASA) and Conference of Robotics and Mechatronics (RobMech)
Abbreviated title	PRASA-RobMech 2017
Country/Territory	South Africa
City	Bloemfontein
Period	29/11/17 → 1/12/17

Access to Document

10.1109/RoboMech.2017.8261139

https://ieeexplore.ieee.org/abstract/document/8261139

Cite this

@inproceedings{feb76ba62ceb419f96b2f22c0efe6722,

title = "Slip estimation methods for proprioceptive terrain classification using tracked mobile robots",

abstract = "Recent work has shown that proprioceptive measurements such as terrain slip can be used for terrain classification. This paper investigates the suitability of four simple slip estimation methods for differentiating between indoor and outdoor terrain surfaces, namely: rocks, grass, rubber and carpet. These slip estimates are calculated using experimental odometric data collected from a tracked autonomous ground vehicle and comprise of two instantaneous estimators and a temporal windowing approach. Results show that only the temporal windowing approach shows significant differences across the terrains investigated, indicating that instantaneous measurements are unsuited to terrain classification.",

author = "Ditebogo Masha and Michael Burke and Bhekisipho Twala",

year = "2018",

month = jan,

day = "18",

doi = "10.1109/RoboMech.2017.8261139",

language = "English",

isbn = "978-1-5386-2315-2",

pages = "150--152",

booktitle = "2017 Pattern Recognition Association of South Africa and Robotics and Mechatronics",

publisher = "Institute of Electrical and Electronics Engineers",

address = "United States",

note = "Symposium of the Pattern Recognition Association of South Africa (PRASA) and Conference of Robotics and Mechatronics (RobMech), PRASA-RobMech 2017 ; Conference date: 29-11-2017 Through 01-12-2017",

}

Masha, D, Burke, M & Twala, B 2018, Slip estimation methods for proprioceptive terrain classification using tracked mobile robots. in 2017 Pattern Recognition Association of South Africa and Robotics and Mechatronics . Institute of Electrical and Electronics Engineers, Bloemfontein, South Africa, pp. 150-152, Symposium of the Pattern Recognition Association of South Africa (PRASA) and Conference of Robotics and Mechatronics (RobMech), Bloemfontein, South Africa, 29/11/17. https://doi.org/10.1109/RoboMech.2017.8261139

Slip estimation methods for proprioceptive terrain classification using tracked mobile robots. / Masha, Ditebogo; Burke, Michael; Twala, Bhekisipho .
2017 Pattern Recognition Association of South Africa and Robotics and Mechatronics . Bloemfontein, South Africa: Institute of Electrical and Electronics Engineers, 2018. p. 150-152.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Slip estimation methods for proprioceptive terrain classification using tracked mobile robots

AU - Masha, Ditebogo

AU - Burke, Michael

AU - Twala, Bhekisipho

PY - 2018/1/18

Y1 - 2018/1/18

N2 - Recent work has shown that proprioceptive measurements such as terrain slip can be used for terrain classification. This paper investigates the suitability of four simple slip estimation methods for differentiating between indoor and outdoor terrain surfaces, namely: rocks, grass, rubber and carpet. These slip estimates are calculated using experimental odometric data collected from a tracked autonomous ground vehicle and comprise of two instantaneous estimators and a temporal windowing approach. Results show that only the temporal windowing approach shows significant differences across the terrains investigated, indicating that instantaneous measurements are unsuited to terrain classification.

AB - Recent work has shown that proprioceptive measurements such as terrain slip can be used for terrain classification. This paper investigates the suitability of four simple slip estimation methods for differentiating between indoor and outdoor terrain surfaces, namely: rocks, grass, rubber and carpet. These slip estimates are calculated using experimental odometric data collected from a tracked autonomous ground vehicle and comprise of two instantaneous estimators and a temporal windowing approach. Results show that only the temporal windowing approach shows significant differences across the terrains investigated, indicating that instantaneous measurements are unsuited to terrain classification.

U2 - 10.1109/RoboMech.2017.8261139

DO - 10.1109/RoboMech.2017.8261139

M3 - Conference contribution

SN - 978-1-5386-2315-2

SP - 150

EP - 152

BT - 2017 Pattern Recognition Association of South Africa and Robotics and Mechatronics

PB - Institute of Electrical and Electronics Engineers

CY - Bloemfontein, South Africa

T2 - Symposium of the Pattern Recognition Association of South Africa (PRASA) and Conference of Robotics and Mechatronics (RobMech)

Y2 - 29 November 2017 through 1 December 2017

ER -

Slip estimation methods for proprioceptive terrain classification using tracked mobile robots

Abstract

Conference

Access to Document

Fingerprint

Cite this