Vehicle-terrain Interaction Research Articles

Terramechanics Models for Tracked Vehicle Terrain Interaction Analysis A Review

Efficient manoeuvrability of off-road tracked vehicles such as military tanks and rovers is essential in ensuring the success of military/extra-terrestrial operations. To achieve this, in-depth research on vehicle-terrain interaction is crucial. This manuscript deals with reviewing the ways to study terramechanics viz. theoretical, empirical, and field tests, and proposing the merits and demerits of each method. Under the theoretical approach, empirical, numerical, and semi-empirical methods are discussed. Under the empirical approach, the method based onthe vehicle cone index for tracked and wheeled vehicles is discussed. Under the numerical approach, advantages and disadvantages of Finite Element Method (FEM) and Discrete Element Method (DEM) are discussed. Semi-empirical method, based upon a combination of the best features of numerical and empirical approaches discusses terrain response to normal repetitive loads and shear repetitive loads for tracked as well as wheeled vehicles. Pressure sinkage relationship for terrains at various loading conditions and shear stress displacement relationship for different terrains obtained through penetration and shear tests are discussed to determine the vehicle’s mobility parameters under a semi-empirical approach. Further, the Super element model, multi-body simulation model, and ride and cornering vibration model are discussed under computer simulation models. A detailed review of various models customized towards tracked vehicle-terrain interaction discussed in this manuscript helps the authors set up a laboratory for terramechanics at DIAT. Preliminary analysis along with conceptual design of the experimental setup is also discussed. In a nutshell, this paper attempts to summarize the research that has been carried out in the field of tracked vehicle-terrain inter action comprising of VCI, MMP, FEM, DEM, Super element model, Multibody technique, and Semi-empirical methods helping the authors to establish a laboratory of terramechanics for their M. Tech. program on Armament and Combat Vehicles at DIAT Pune.

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

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.

Real-Time Simulation of Ground Vehicles on Deformable Terrain

Abstract Sophisticated modeling and simulation, based on rigid and flexible multibody dynamics, are nowadays a standard procedure in the design and analysis of vehicle systems and are widely adopted for on-road driving. Off-road driving for both terrestrial wheeled and tracked vehicles, as well as wheeled and legged robots and rovers for extra-terrestrial exploration pose additional modeling and simulation challenges, a primary one being that of the vehicle–terrain interaction, modeling of deformable terrain, and terramechanics in general. Techniques for modeling deformable terrain span an entire range varying in complexity, representation accuracy, and ensuing computational effort. While formulations such as fully resolved granular dynamics, continuum representation of granular material, or finite element can provide a high level of accuracy, they do so at a significant cost, even when the implementation leverages parallel computing and/or hardware accelerators. Real-time or faster than real-time terramechanics is a highly desired capability (in applications such as training of autonomous vehicles and robotic systems) or critical capability (in applications such as human-in-the-loop or hardware-in-the-loop). We present a real-time capable deformable soil implementation, extended from the soil contact model (SCM) developed at the German Aerospace Center which in turn can be viewed as a generalization of the Bekker-Wong and Janosi-Hanamoto semi-empirical models for soil interaction with arbitrary three-dimensional shapes and arbitrary contact patches. This SCM implementation is available, alongside more computationally intensive deformable soil representations, in the open-source multiphysics package Chrono. We describe the overall implementation and the features of the Chrono SCM model, the efficient underlying data structures, the current multicore parallelization aspects, and its scalability properties for concurrent simulation of multiple vehicles on deformable terrain.

ScaTE: A Scalable Framework for Self- Supervised Traversability Estimation in Unstructured Environments

For the safe and successful navigation of autonomous vehicles in unstructured environments, the traversability of terrain should vary based on the driving capabilities of the vehicles. Actual driving experience can be utilized in a self-supervised fashion to learn vehicle-specific traversability. However, existing methods for learning self-supervised traversability are not highly scalable for learning the traversability of various vehicles. In this work, we introduce a scalable framework for learning self-supervised traversability, which can learn the traversability directly from vehicle-terrain interaction without any human supervision. We train a neural network that predicts the proprioceptive experience that a vehicle would undergo from 3D point clouds. Using a novel PU learning method, the network simultaneously identifies non-traversable regions where estimations can be overconfident. With driving data of various vehicles gathered from simulation and the real world, we show that our framework is capable of learning the self-supervised traversability of various vehicles. By integrating our framework with a model predictive controller, we demonstrate that estimated traversability results in effective navigation that enables distinct maneuvers based on the driving characteristics of the vehicles. In addition, experimental results validate the ability of our method to identify and avoid non-traversable regions.

Vehicle design for terrain mobility: A modeling technique of powertrain power conversion and realization

Vehicle terrain mobility characteristics, provided by the powertrain and running gear, are realized in dynamic interactions between the wheels and terrain. Approaches to modeling and simulation of vehicle-terrain interaction and mobility characteristics as well as engineering approaches to design powertrain sub-systems together pre-determine a vehicle’s technical success or failure before it touches the ground. This article develops a vehicle mobility design technique, applicable to both manned and unmanned platforms, concerned with powertrain power conversion and realization in tire-terrain interactions. The modeling component is based on multi-drive-wheel vehicle longitudinal dynamics combined with terramechanics and powertrain characteristics. The approach advances the conventional dynamic factor by introducing the conjoint effect of the engine, transmission, and driveline system on vehicle traction and acceleration performance in terrain conditions where circumferential wheel forces and tire slippages may differ from each other. The vehicle design component of the proposed technique introduces drivetrain, driveline, and powertrain design factors that assess the influence of the drivetrain and driveline systems on traction, acceleration performance, power conversion, and realization at the wheels. The vehicle-design-for-mobility technique is completed by examining indices of mobility margins and performance. An analysis of several 8x8 armored personal carriers and 4x4 off-road vehicles illustrates the proposed technique.

Development of sprung mass non-linear mathematical model of pitch dynamics with trailing arm suspension

Military vehicles are required to negotiate drastic ride environments. Therefore, the vibration magnitudes which are transmitted to the vehicle chassis through the suspension system due to dynamic vehicle-terrain interactions, are usually large. Hence, it is necessary to study the vibrations that are transmitted to the sprung mass as it negotiates different types of terrain at different speeds. The present study is focused on the development of sprung mass non-linear pitch dynamics mathematical model of a military vehicle with a trailing arm torsion bar suspension system. The trailing arm kinematics and dynamics are incorporated into the non-linear governing equations of motion of the sprung and unsprung masses which consider the effects of sprung mass large pitch angular motions. The model is solved by coding in Matlab. The sprung mass non-linear pitch dynamics mathematical model is validated with an equivalent multi-body dynamics model which is developed by using MSC.ADAMS. This mathematical model would play a key role to fine-tune the vehicle and suspension parameters as well as comparatively evaluate the performance of the hydro-gas suspension system. The model can further be extended to include the military vehicle weapon recoil effects which can cause considerable magnitudes of vehicle pitch. Apart from simulating ride dynamics of the entire vehicle, the influence of movement with crane payload on the military recovery vehicle pitch dynamics can also be brought out as a useful extension of this mathematical model.

On-Line Learning and Updating Unmanned Tracked Vehicle Dynamics

Increasing levels of autonomy impose more pronounced performance requirements for unmanned ground vehicles (UGV). Presence of model uncertainties significantly reduces a ground vehicle performance when the vehicle is traversing an unknown terrain or the vehicle inertial parameters vary due to a mission schedule or external disturbances. A comprehensive mathematical model of a skid steering tracked vehicle is presented in this paper and used to design a control law. Analysis of the controller under model uncertainties in inertial parameters and in the vehicle-terrain interaction revealed undesirable behavior, such as controller divergence and offset from the desired trajectory. A compound identification scheme utilizing an exponential forgetting recursive least square, generalized Newton–Raphson (NR), and Unscented Kalman Filter methods is proposed to estimate the model parameters, such as the vehicle mass and inertia, as well as parameters of the vehicle-terrain interaction, such as slip, resistance coefficients, cohesion, and shear deformation modulus on-line. The proposed identification scheme facilitates adaptive capability for the control system, improves tracking performance and contributes to an adaptive path and trajectory planning framework, which is essential for future autonomous ground vehicle missions.

Application of Two Techniques Used for Measuring the Soil Strength : A Review

This paper reviews the application of two techniques are used for the measurement of soil strength. The two techniques (cone penetrometer and bevameter) enable measurements to be made in situ. However, the penetrometer is the only equipment that can evaluate soil resistance variation with depth. Neither the cone index nor the gradient that belongs to it concerning the depth is uniquely associated with density or soil cohesion but varies with structural state and moisture content. Moreover, the formation of compaction zones and soil bodies ahead of the cone efficiently alters its geometry, thus the penetration no longer reveals the original characteristics of the soil. Furthermore, the bevameter technique only identifies surface soil properties. Nonetheless, among all the presently available techniques, the bevameter gives the closest imitation of vehicle-terrain interaction.

Self-Supervised Visual Terrain Classification From Unsupervised Acoustic Feature Learning

Mobile robots operating in unknown urban environments encounter a wide range of complex terrains to which they must adapt their planned trajectory for safe and efficient navigation. Most existing approaches utilize supervised learning to classify terrains from either an exteroceptive or a proprioceptive sensor modality. However, this requires a tremendous amount of manual labeling effort for each newly encountered terrain as well as for variations of terrains caused by changing environmental conditions. In this article, we propose a novel terrain classification framework leveraging an unsupervised proprioceptive classifier that learns from vehicle-terrain interaction sounds to self-supervise an exteroceptive classifier for pixelwise semantic segmentation of images. To this end, we first learn a discriminative embedding space for vehicle-terrain interaction sounds from triplets of audio clips formed using visual features of the corresponding terrain patches and cluster the resulting embeddings. We subsequently use these clusters to label the visual terrain patches by projecting the traversed tracks of the robot into the camera images. Finally, we use the sparsely labeled images to train our semantic segmentation network in a weakly supervised manner. We present extensive quantitative and qualitative results that demonstrate that our proprioceptive terrain classifier exceeds the state-of-the-art among unsupervised methods and our self-supervised exteroceptive semantic segmentation model achieves a comparable performance to supervised learning with manually labeled data.

Parallelized Multiscale Off-Road Vehicle Mobility Simulation Algorithm and Full-Scale Vehicle Validation

Abstract In this work, a scalable parallel computing scheme for the hierarchical multiscale off-road vehicle mobility simulation capability is developed with the hybrid message passing interface (MPI)/OpenMP framework, and it is validated against full-scale vehicle test data. While the hierarchical multiscale modeling approach has been introduced to high-fidelity off-road mobility simulations to eliminate limitations of existing single-scale deformable terrain models, computational complexities associated with the large dimensionality of multibody vehicle equations, involving nonlinear finite element tires and multiscale terrain models, need to be addressed for use in full-scale vehicle mobility predictions. To this end, a co-simulation framework for the multiscale off-road vehicle mobility model is proposed by exploiting the moving soil patch technique. This allows for systematically extracting four moving soil patches for four tires in a vehicle model from the deformable terrain domain, and time integrations of the four tire–soil subsystems are performed concurrently to enable computational speedup. Furthermore, an automated updating scheme for multiscale moving soil patches for a full vehicle model is developed, considering the multipass effect in various vehicle maneuvering scenarios. To demonstrate the off-road mobility prediction capability using the proposed parallelized multiscale vehicle–terrain interaction simulation algorithm, full-scale vehicle validation is presented for the vehicle drawbar pull as well as variable grade hill climb tests on soft soil.

Improving Limitations of Rover Missions in the Moon and Planets by Unifying Vehicle–Terrain Interaction Models

This paper extends the analysis of empirical methods for describing vehicle–terrain interactions in lunar terrain. Given analytical formulation to predict mobility performance in extraterrestrial environments requires reliable in situ testing campaigns, this imposes fundamental restrictions to conceive a more consolidated theory, and further any possibility to use improved empirical methods to design better space hardware. Hence, we propose an analytical approach to extrapolate data taken in parabolic flights to model vehicle performance in multiple gravity regimes. The extrapolation technique and respective reported uncertainties can be used, therefore, to tune fitting parameters of a set of general formulas in the domain of Terramechanics, allowing to have empirical estimates of the wheel mobility in cases where testing is inherently very complex. Finally, by the analysis of previous Moon and Mars rover missions with the fitted equations, we report an empirical design criterion that allows to generate first estimates of the optimal wheel dimensions, taking into account the eventual longitudinal slip of the rover and including the desired mass of the payload. The introduced rules for geometry optimization can be important for future space rover missions in remote soils.

Testing Design Optimization for Uncertainty Reduction in Generating Off-Road Mobility Map Using a Bayesian Approach

Abstract The Next Generation NATO Reference Mobility Model (NG-NRMM) plays a vital role in vehicle mobility prediction and mission planning. The complicated vehicle–terrain interactions and the presence of heterogeneous uncertainty sources in the modeling and simulation (M&S) result in epistemic uncertainty/errors in the vehicle mobility prediction for given terrain and soil conditions. In this paper, the uncertainty sources that cause the uncertainty in mobility prediction are first partitioned into two levels, namely uncertainty in the M&S and uncertainty in terrain and soil maps. With a focus on the epistemic uncertainty in the M&S, this paper presents a testing design optimization framework to effectively reduce the uncertainty in the M&S and thus increase the confidence in generating off-road mobility maps. A Bayesian updating approach is developed to reduce the epistemic uncertainty/errors in the M&S using mobility testing data collected under controllable terrain and soil conditions. The updated models are then employed to generate the off-road mobility maps for any given terrain and soil maps. Two types of design strategies, namely testing design for model selection and testing design for uncertainty reduction, are investigated in the testing design framework to maximize the information gain subject to limited resources. Results of a numerical example demonstrate the effectiveness of the proposed mobility testing design optimization framework.

Hierarchical Multiscale Modeling of Tire–Soil Interaction for Off-Road Mobility Simulation

A high-fidelity computational terrain dynamics model plays a crucial role in accurate vehicle mobility performance prediction under various maneuvering scenarios on deformable terrain. Although many computational models have been proposed using either finite element (FE) or discrete element (DE) approaches, phenomenological constitutive assumptions in FE soil models make the modeling of complex granular terrain behavior very difficult and DE soil models are computationally intensive, especially when considering a wide range of terrain. To address the limitations of existing deformable terrain models, this paper presents a hierarchical FE–DE multiscale tire–soil interaction simulation capability that can be integrated in the monolithic multibody dynamics solver for high-fidelity off-road mobility simulation using high-performance computing (HPC) techniques. It is demonstrated that computational cost is substantially lowered by the multiscale soil model as compared to the corresponding pure DE model while maintaining the solution accuracy. The multiscale tire–soil interaction model is validated against the soil bin mobility test data under various wheel load and tire inflation pressure conditions, thereby demonstrating the potential of the proposed method for resolving challenging vehicle-terrain interaction problems.

Cosimulation of vehicle dynamics and terrain interaction to predict one-pass vehicle cone index

The one-pass vehicle cone index (VCI1) is a measurement of the softest soil over which a vehicle can consistently drive at least once. In the last few decades, VCI1 has become a common quantitative indicator for the soft-soil performance of wheeled and tracked ground vehicles. Because VCI1 depends on not only the strength (as measured by Cone Index) of the soil but also the type of the soil, it is difficult to numerically predict VCI1 for notional ground vehicles. While reliable empirical methods to predict VCI1 have been refined over the years, the emergence of new vehicle technologies such as traction control systems presents a new challenge for ground vehicle modelling and necessitates a new way for predicting VCI1 performance. In this paper, we present accurate numerical predictions of VCI1 that use the Chrono multi-body dynamics code for vehicle dynamics, cosimulated with empirical, dynamic traction predictions at the vehicle-terrain interface by using the Ground Contact Element (GCE) code. We show that this cosimulated approach accurately reproduces measured values for a variety of wheeled vehicles.

Power consumption of tracked and wheeled small mobile robots on deformable terrains–model and experimental validation

The ability to predict the power consumption of a small mobile robot with either tracked or wheeled configuration is important to its design and control. Accurate and fast prediction of power consumption enables online motion planning that does not exceed the power limits imposed by the propulsion system or the terrain. Skid steering is widely used on small mobile robots because of its simplicity and robust performance. However, modeling of skid steering on deformable terrains is difficult because of the complex vehicle-terrain interaction and the distributed nature of shear stress along the contact surface. This paper first develops a new skid steering model based on terramechanics that is capable of predicting torque and power consumption and is both accurate and computationally efficient. This skid steering model is verified using the experimental data obtained from a track/wheel interchangeable mobile robot driving on dry sand at low speeds. Further, this paper studies the comparison of power consumption between tracked and wheeled mobile robots on deformable terrains focusing on two different aspects: 1) the power losses due to internal friction and damping resistances of both tracked and wheeled configurations, measured from the robot, and 2) the power consumed by deforming the soft soil, using the power consumption model. Overall, this study provides new insights on power consumption of skid steering of both tracked and wheeled mobile robots on deformable terrains.

In brief

PAGE 582 Researchers in Germany present a red emitting laser with V-shaped surface grooves in the p-confinement layer, which result in low radiation losses. The laser will be suitable for many applications that require laser sources that emit high-power at precisely defined wavelengths, such as interferometry, free-space communication, metrology and light detection. PAGE 558 A collision avoidance scheme for ground vehicles is proposed in work from Korea. The scheme including a modified turn algorithm improves driving performance by minimising sudden curvature changes and side-slips of the vehicle. A non-linear estimator for time-to-collision is also introduced to increase collision avoidance. Surface grooves in a red emitting laser result in low radiation losses to produce high output power. PAGE 584 Work from the US demonstrates that a 3D metamaterial structure with back-to-back, vertically oriented metallic cross dipoles efficiently linearly polarises light. The control over light polarisation in the presented metamaterial surpasses that of traditional 2D polarisers due to a reduction in reflected and transmitted polarisation. A scheme to improve driving performance reduces sudden changes in curvature of ground vehicles. PAGE 580 Researchers in Japan design an all-optical AND logic gate using photonic crystal (PC) quantum dot semiconductor optical amplifiers (QDSOA) for optical equipment, like multiplexers. The proposed gate is compared to traditional QDSOA optical gates and the incorporation of a PC is shown to reduce energy consumption and device volume. Efficient linear polarisation of light with a 3D metamaterial structure. PAGE 595 A method to resolve terrain features to adjust the trajectory of a mobile robot using vibrations from vehicle-terrain interactions is reported in work from Korea. The presented algorithm allows a robot that explores an unfamiliar area to efficiently plan their path across different surfaces with differing features and textures. An all-optical quantum dot semiconductor photonic crystal AND logic gate is presented with low energy consumption. Vehicle-terrain interactions are used to adjust a planned route for efficient terrain exploration.

MRF‐based terrain map inference using variational feature projection

A method to infer the terrain of unobserved fields as well as navigation paths using vibrations from vehicle–terrain interactions is presented. The relationships between adjacent terrain grids and robot paths are modelled by Markov random field (MRF) for inferring the classes of unobserved field. In order to approximate the complex likelihood of vibration features as a tractable distribution, a variational auto-encoder is adopted. By projecting intractable features to the latent space, an approximated likelihood that gives similar performance to a neural network classifier is obtained. Then the numerical maximum a posterior inference for MRF is obtained. The result of inferred terrain maps are compared with that from the feature likelihood as a Gaussian-mixture model.

Study of vehicle-terrain interaction based on ANCF

A brief review of the empirical and analytical models are presented. The pressure-sinkage relationship proposed by Bekker and the tractive performance (drawbar pull-slip) of vehicle is set up. A figure of drawbar pull vs. slip at different time and slip is illustrated. The modified Bekker soil models are integrated finite element (FE) Absolute Nodal Coordinate Formulation (ANCF) interpolations to determine the generalized forces. The paper discusses important fundamental issues that must be addressed when implementing terramechanics models into MBS algorithms in order to model complex tracked vehicle-soil interactions.

Deep spatiotemporal models for robust proprioceptive terrain classification

Terrain classification is a critical component of any autonomous mobile robot system operating in unknown real-world environments. Over the years, several proprioceptive terrain classification techniques have been introduced to increase robustness or act as a fallback for traditional vision based approaches. However, they lack widespread adaptation due to various factors that include inadequate accuracy, robustness and slow run-times. In this paper, we use vehicle-terrain interaction sounds as a proprioceptive modality and propose a deep long-short term memory based recurrent model that captures both the spatial and temporal dynamics of such a problem, thereby overcoming these past limitations. Our model consists of a new convolution neural network architecture that learns deep spatial features, complemented with long-short term memory units that learn complex temporal dynamics. Experiments on two extensive datasets collected with different microphones on various indoor and outdoor terrains demonstrate state-of-the-art performance compared to existing techniques. We additionally evaluate the performance in adverse acoustic conditions with high-ambient noise and propose a noise-aware training scheme that enables learning of more generalizable models that are essential for robust real-world deployments.

Dynamics modeling and robotic-assist, leader-follower control of tractor convoys

This paper proposes a generalized dynamics model and a leader-follower control architecture for skid-steered tracked vehicles towing polar sleds. The model couples existing formulations in the literature for the powertrain components with the vehicle-terrain interaction to capture the salient features of terrain trafficability and predict the vehicles response. This coupling is essential for making realistic predictions of the vehicles traversing capabilities due to the power-load relationship at the engine output. The objective of the model is to capture adequate fidelity of the powertrain and off-road vehicle dynamics while minimizing the computational cost for model based design of leader-follower control algorithms. The leader-follower control architecture presented proposes maintaining a flexible formation by using a look-ahead technique along with a way point following strategy. Results simulate one leader-follower tractor pair where the leader is forced to take an abrupt turn and experiences large oscillations of its drawbar arm indicating potential payload instability. However, the follower tractor maintains the flexible formation but keeps its payload stable. This highlights the robustness of the proposed approach where the follower vehicle can reject errors in human leader driving.