Wheel-terrain interaction plays a critical role for vehicle mobility on natural terrain, such as in agricultural, planetary exploration and off-road settings. Estimation of the terrain characteristics and the way they affect traversability is essential for the vehicle to better plan its safest and energy-efficient path. This work proposes a novel approach to learn and predict from a distance the motion resistance encountered by a robotic vehicle, while traversing natural soil, by using visual information from a stereovision device. To this end, terrain appearance and geometry information are first correlated to resistance torque measurements during a learning phase via two alternative regression approaches, namely Least-Squares Boosting and Long-Short Term Memory Recurrent Neural Network. Then, such a relationship is exploited to predict motion resistance remotely, based on visual data only. Results obtained in preliminary experimental tests on ploughed and compact terrain are presented to show the feasibility of the proposed method.

Learning and prediction of vehicle-terrain interaction from 3D vision / Ugenti, A.; Vulpi, F.; Milella, A.; Reina, G.. - STAMPA. - 11785:(2021). (Intervento presentato al convegno Multimodal Sensing and Artificial Intelligence: Technologies and Applications II 2021 tenutosi a deu nel 2021) [10.1117/12.2596021].

Learning and prediction of vehicle-terrain interaction from 3D vision

Vulpi F.;Reina G.
Conceptualization
2021-01-01

Abstract

Wheel-terrain interaction plays a critical role for vehicle mobility on natural terrain, such as in agricultural, planetary exploration and off-road settings. Estimation of the terrain characteristics and the way they affect traversability is essential for the vehicle to better plan its safest and energy-efficient path. This work proposes a novel approach to learn and predict from a distance the motion resistance encountered by a robotic vehicle, while traversing natural soil, by using visual information from a stereovision device. To this end, terrain appearance and geometry information are first correlated to resistance torque measurements during a learning phase via two alternative regression approaches, namely Least-Squares Boosting and Long-Short Term Memory Recurrent Neural Network. Then, such a relationship is exploited to predict motion resistance remotely, based on visual data only. Results obtained in preliminary experimental tests on ploughed and compact terrain are presented to show the feasibility of the proposed method.
2021
Multimodal Sensing and Artificial Intelligence: Technologies and Applications II 2021
Learning and prediction of vehicle-terrain interaction from 3D vision / Ugenti, A.; Vulpi, F.; Milella, A.; Reina, G.. - STAMPA. - 11785:(2021). (Intervento presentato al convegno Multimodal Sensing and Artificial Intelligence: Technologies and Applications II 2021 tenutosi a deu nel 2021) [10.1117/12.2596021].
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11589/272990
Citazioni
  • Scopus 1
  • ???jsp.display-item.citation.isi??? 1
social impact