DeepTerramechanics: Terrain Classification and Slip Estimation for Ground Robots via Deep Learning

Gonzalez, Ramon; Iagnemma, Karl

doi:10.48550/arxiv.1806.07379

Cited by 5 publications

(8 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The traditional path planners, as noted in prior research, are designed for operation within structured and homogeneous environments, meaning they do not require explicit consideration of the vehicle-terrain interaction [16]. However, recent studies on TTA have underscored the critical significance of vehicle-terrain interaction [42][43][44]. This interaction can give rise to variations such as wheel slip, wheel entrapment, and vehicle rollover, which consequently influence the overall traversability of the terrain.…”

Section: Vehicle-terrain Interactionmentioning

confidence: 99%

A Survey on Path Planning for Autonomous Ground Vehicles in Unstructured Environments

Wang,

Li,

Zhang

et al. 2024

Machines

View full text Add to dashboard Cite

Autonomous driving in unstructured environments is crucial for various applications, including agriculture, military, and mining. However, research in unstructured environments significantly lags behind that in structured environments, mainly due to the challenges posed by harsh environmental conditions and the intricate interactions between vehicles and terrains. This article first categorizes unstructured path planning into hierarchical and end-to-end approaches and then the special parts compared to structured path planning are emphatically reviewed, such as terrain traversability analysis, cost estimation, and terrain-dependent constraints. This article offers a comprehensive review of the relevant factors, vehicle–terrain interactions, and methods of terrain traversability analysis. The estimation methods of safety cost, energy cost, and comfort cost are also emphatically summarized. Moreover, the constraints caused by the limits of terrains and vehicles are discussed. The applications of algorithms in recent articles for path planners are reviewed. Finally, crucial areas requiring further research are analyzed in unstructured path planning.

show abstract

Section: Vehicle-terrain Interactionmentioning

confidence: 99%

A Survey on Path Planning for Autonomous Ground Vehicles in Unstructured Environments

Wang,

Li,

Zhang

et al. 2024

Machines

View full text Add to dashboard Cite

show abstract

“…Detection of the slip has been addressed in several places using exteroceptive sensors (radar, Lidar, Global Positioning System (GPS), camera), proprioceptive senors (Inertial Measurement Unit (IMU), encoder, gyrometer), or a combination of them. In [1], the slip detection methods for planetary WMRs are categorized into direct signal-based [21], [22], estimation-based [23], [24], [25], [26], [27], [28], [29], terramechanic-based [30], [31], machine learning [32], [33], [34], [35], [36], and global sensing [37], [38] approaches. Most of the estimation-based approaches are sensor-level fusions using EKF or UKF techniques to provide systematic solutions.…”

Section: B Slip Estimationmentioning

confidence: 99%

Sequential Track Fusion in UKF-based Multi-Sensor Networks: A Case of Slip Estimation in Wheeled Mobile Robots

Zarei¹,

Chhabra²,

Mottaghi³

2023

Preprint

View full text Add to dashboard Cite

Accurate wheel slip estimation facilitates Wheeled Mobile Robots (WMRs) with improved localization and traversability monitoring which are crucial to their autonomy in challenging environments. Although distributed track-level fusion offers better computational efficiency, often sensor-level fusion is adopted in slip estimators. Extending upon our earlier work, we develop a novel explicit track-to-track fusion algorithm for UKF-based multi-sensor networks that has immediate application to slip ratio estimation in WMRs. The algorithm demonstrates better consistency compared to the existing fusion techniques, since it implements an optimal fusion rule to sequentially combine local tracks. It also saves computational power due to the recursive propagation of cross-covariance matrices based on the statistical linearization technique, instead of performing online optimizations. We prove that this recursion only requires the information of the first and last tracks in a sequence if all local tracks are unbiased. We rigorously study various key properties of the developed fusion algorithm and show its superior level of confidence when compared to the two prominent fusion methods of sequential and batch covariance intersection. The slip ratio estimation in a six-wheel WMR is considered as a case study, where the steerable wheel sets act as a network of sensors. The proposed slip estimator works based on the rigid body kinematics and readings of purely proprioceptive sensors, i.e., an inertial measurement unit and encoders. The slip estimator's performance is evaluated in a high-fidelity software-in-the-loop simulation environment connecting MATLAB and CM Lab's Vortex Studio software. In a comparison study that considers four rival strategies, we show the superiority of the proposed fusion method offering a balance between consistency, accuracy, and speed in real-time slip estimations.

show abstract

“…Thirdly, a solution to the joint vehicle state and parameter estimation problem using CKF is proposed by augmenting the state vector of the filter with the equivalent soil stiffness. Other approaches have attempted to estimate terrain properties by collecting various sets of features through the onboard sensor suite that are used to train a ground classifier that uses a machine learning algorithm including Support Vector Machine (SVM) (Ward and Iagnemma 2009;Reina et al 2017a), Bayesian network (Galati and Reina 2019), and Deep Learning (Gonzalez and Iagnemma 2018). Here, a model-based approach is used where a general vertical dynamic model is adopted.…”

Section: Related Researchmentioning

confidence: 99%

Terrain estimation via vehicle vibration measurement and cubature Kalman filtering

Reina

Leanza

Messina

2019

Journal of Vibration and Control

View full text Add to dashboard Cite

The extent of vibrations experienced by a vehicle driving over natural terrain defines its ride quality. Generally, surface irregularities, ranging from single discontinuities to random variations of the elevation profile, act as a major source of excitation that induces vibrations in the vehicle body through the tire-soil interaction and suspension system. Therefore, the ride response of off-road-vehicles is tightly connected with the ground properties. The objective of this research is to develop a model-based observer that estimates automatically terrain parameters using available onboard sensors. Two acceleration signals, one coming from the vehicle body and one from the wheel suspension are fed into a dynamic vehicle model that takes into account tire/terrain interaction to estimate ground properties. To solve the resulting nonlinear simultaneous state and parameter estimation problem, the Cubature Kalman filter is used that is shown to outperform the standard Extended Kalman filter in terms of accuracy and stability. An extensive set of simulation tests is presented to assess the performance of the proposed estimator under various surface roughness and deformability conditions. Results show the potential of the proposed observer to estimate automatically terrain properties during operations that could be implemented onboard of a general family of intelligent vehicles, ranging from off-road high-speed passenger cars to lightweight and low-speed planetary rovers. Prepared using sagej.cls arXiv:2001.05165v1 [eess.SY] 15 Jan 2020

show abstract

DeepTerramechanics: Terrain Classification and Slip Estimation for Ground Robots via Deep Learning

Cited by 5 publications

References 42 publications

A Survey on Path Planning for Autonomous Ground Vehicles in Unstructured Environments

A Survey on Path Planning for Autonomous Ground Vehicles in Unstructured Environments

Sequential Track Fusion in UKF-based Multi-Sensor Networks: A Case of Slip Estimation in Wheeled Mobile Robots

Terrain estimation via vehicle vibration measurement and cubature Kalman filtering

Contact Info

Product

Resources

About