Autonomous Ground Vehicles Research Articles

R2-RRT*: Reliability-Based Robust Mission Planning of Off-Road Autonomous Ground Vehicle Under Uncertain Terrain Environment

This article presents a reliable and robust rapidly exploring random tree (R2-RRT&#x002A;) algorithm to tackle challenges in mission planning of off-road autonomous ground vehicles (AGVs) under uncertain terrain environment. Two types of mobility reliability metrics, namely state mobility reliability (SMR) and mission mobility reliability (MMR), are first defined to quantify the mobility reliability of an AGV and to incorporate mobility reliability into mission planning. SMR measures the probability that a vehicle can pass through a specific location on a map of interest, whereas MMR quantifies mobility reliability of a mission path with the consideration of dependence of soil properties and slope over space. Based on the defined SMR and MMR metrics, two reliability-based robust mission planning models are developed to identify optimal paths that have robust travel time and satisfy specific reliability requirements. Moreover, a reliability-based path smoothing algorithm is developed to address the suboptimality of R2-RRT&#x002A;. Results of a case study demonstrate the efficacy of the proposed models and algorithms. <i>Note to Practitioners</i>&#x2014;This article was motivated to explicitly account for the uncertain terrain environment in mission planning of off-road autonomous ground vehicles (AGVs). Existing approaches, e.g., RRT and its variants, in general, oversimplify the uncertainty sources and overlook the reliability of vehicle mobility. These simplifications could lead to failure (i.e., immobility) of off-road AGVs on the obtained paths. This article suggests a reliability-based mission planning model by incorporating the proposed SMR and MMR metrics into mission planning considering the spatial-dependent uncertainty sources. To identify a reliable and robust path, we extend RRT&#x002A; to R2-RRT&#x002A; that achieves a tradeoff between mission cost and reliability. The proposed reliability-based mission planning model, however, is not limited to RRT&#x002A; and can also be integrated with other path planning algorithms. It can be also applied to other unmanned vehicles and robots, such as the motion planning of unmanned aerial vehicles (UAVs) in adverse weather conditions.

Piecewise Linear Path Following for a Unicycle Using Transverse Feedback Linearization

Motion planning and path following together constitute a critical part of the decision-making hierarchy in autonomous ground vehicles. One of the simplest instances of this architecture is when the path planner generates waypoints that define a sequence of collision-free line segments from a start location to goal destination and when the vehicle’s kinematic model is taken to be Dubin’s vehicle. The low-level feedback controller can then be designed by treating the path following problem as a set stabilization problem; one such approach is called transverse feedback linearization (TFL). However, for a Dubin’s vehicle with only one input, the direction of traversal along the path is completely determined by the vehicle’s initial condition. In this article, we provide easily checkable sufficient conditions that only need to be checked once at run time and a simple systematic design procedure, which guarantees that the robot traverses the path while coming within a user-specified distance of the waypoints; the procedure is also empirically verified.

Trajectory Tracking Control of Autonomous Ground Vehicles Using Adaptive Learning MPC.

In this work, an adaptive learning model predictive control (ALMPC) scheme is proposed for the trajectory tracking of perturbed autonomous ground vehicles (AGVs) subject to input constraints. In order to estimate the unknown system parameter, we propose a set-membership-based parameter estimator based on the recursive least-squares (RLS) technique with the ensured nonincreasing estimation error. Then, the estimated system parameter is employed in MPC to improve the prediction accuracy. In the proposed ALMPC scheme, a robustness constraint is introduced into the MPC optimization to handle parametric and additive uncertainties. For the designed robustness constraint, its shape is decided off-line based on the invariant set, whereas its shrinkage rate is updated online according to the estimated upper bound of the estimation error, leading to further reduced conservatism and slightly increased computational complexity compared with the robust MPC methods. Furthermore, it is theoretically shown that the proposed ALMPC algorithm is recursively feasible under some derived conditions, and the closed-loop system is input-to-state stable (ISS). Finally, a numerical example and comparison study are conducted to illustrate the efficacy of the proposed method.

Path Planning Based on Improved Hybrid A*Algorithm

Hybrid A*algorithm has been widely used in mobile robots to obtain paths that are collision-free and drivable. However, the outputs of hybrid A*algorithm always contain unnecessary steering actions and are close to the obstacles. In this paper, the artificial potential field (APF) concept is applied to optimize the paths generated by the hybrid A*algorithm. The generated path not only satisfies the non-holonomic constraints of the vehicle, but also is smooth and keeps a comfortable distance to the obstacle at the same time. Through the robot operating system (ROS) platform, the path planning experiments are carried out based on the hybrid A*algorithm and the improved hybrid A*algorithm, respectively. In the experiments, the results show that the improved hybrid A*algorithm greatly reduces the number of steering actions and the maximum curvature of the paths in many different common scenarios. The paths generated by the improved algorithm nearly do not have unnecessary steering or sharp turning before the obstacles, which are safer and smoother than the paths generated by the hybrid A*algorithm for the autonomous ground vehicle.

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.

Takagi–Sugeno Fuzzy Unknown Input Observers to Estimate Nonlinear Dynamics of Autonomous Ground Vehicles: Theory and Real-Time Verification

In this article, we address the simultaneous estimation problem of the lateral speed, the steering input, and the effective engine torque, which play a fundamental role in vehicle handling, stability control, and fault diagnosis of autonomous ground vehicles. Due to the involved longitudinal-lateral coupling dynamics and the presence of unknown inputs (UIs), a new nonlinear observer design technique is proposed to guarantee the asymptotic estimation performance. To this end, we make use of a specific Takagi-Sugeno (TS) fuzzy representation with nonlinear consequents to exactly model the nonlinear vehicle dynamics within a compact set of the vehicle state. This TS fuzzy modeling not only allows reducing significantly the real-time computational effort in estimating the vehicle variables but also enables an effective way to deal with unmeasured nonlinearities. Moreover, via a generalized Luenberger observer structure, the UI decoupling can be achieved without requiring a priori UI information. Using Lyapunov stability arguments, the UI observer design is reformulated as an optimization problem under linear matrix inequalities, which can be effectively solved with standard numerical solvers. The effectiveness of the proposed TS fuzzy UI observer design is demonstrated with real-time hardware-in-the-loop experiments.

Probabilistic Indoor Positioning and Navigation (PIPN) of Autonomous Ground Vehicle (AGV) Based on Wireless Measurements

Recently, Autonomous Ground Vehicles (AGV) and mobile robots have been rapidly developed in various engineering applications, such as Industry 4.0 factory and smart manufacturing. Indoor navigation was one of the most important tasks for the mobile systems as they were often designed to move from one location to another location autonomously without contacting the surrounding objects along the moving path in a usually dynamic and complex indoor environment. There were two key steps to achieve Simultaneous Localization and Mapping (SLAM). First, indoor positioning of the mobile system based on some measurements was done. The second step was to navigate itself inside the indoor map. This was a very challenging problem because there always existed uncertainties in the measurements. It was desired to estimate the positioning errors and determine a safe moving path with high reliability. This paper presented the methodologies for wireless indoor positioning and navigation of AGV with measurement uncertainties. Two kinds of AGV moving trajectories with various design parameters were simulated: a linear trajectory and a curved one. It was found that both greater number of sensors being used for wireless measurements and greater number of measurement trials for multilateration could effectively improve the accuracy of AGV positioning.

Takagi-Sugeno Fuzzy Fault Detector Design with Finite-Frequency Specifications for Autonomous Ground Vehicles

This paper investigates the steering fault detection problem for autonomous ground vehicles (AGVs). Using an observer-based approach, a new fuzzy fault detector for steering actuator is designed for safety concern. To this end, a two degrees-of-freedom (2-DOF) nonlinear vehicle model is adopted to represent the nonlinear dynamics of AGVs. Since it is not easy to measure the lateral velocity in practice, this model is then represented in a specific Takagi-Sugeno (TS) fuzzy form with nonlinear consequents. In contrast to the conventional TS fuzzy modeling, it allows separating the unmeasured premise variables in the local nonlinear consequent, which enables a more effective way to deal with the challenging issue of unmeasured premise variables. Moreover, to minimize the effect of disturbances on system performance and maximize that of actuator faults on the generated residual, both H∞ disturbance attenuation index and H_fault sensitivity index are taken into account in a finite-frequency domain. The conditions to design fault detection TS fuzzy observer are derived using Lyapunov stability method. The design procedure can be reformulated as an optimization problem under linear matrix inequalities, efficiently solved by standard numerical solvers. Simulation results are given to verify the fault detection performance of the proposed method.

Steering Angle Prediction Techniques for Autonomous Ground Vehicles: A Review

Unintentional lane departure accidents are one of the biggest reasons for the causalities that occur due to human errors. By incorporating lane-keeping features in vehicles, many accidents can be avoided. The lane-keeping system operates by auto-steering the vehicle in order to keep it within the desired lane, despite of changes in road conditions and other interferences. Accurate steering angle prediction is crucial to keep the vehicle within the road boundaries, which is a challenging task. The main difficulty in this regard is to identify the drivable road area on heterogeneous road types varying in color, texture, illumination conditions, and lane marking types. This strenuous problem can be addressed by two approaches, namely, `computer-vision-based approach' and `imitation-learning-based approach'. To the best of our knowledge, at present, there is no such detailed review study covering both the approaches and their related optimization techniques. This comprehensive review attempts to provide a clear picture of both approaches of steering angle prediction in the form of step by step procedures. The taxonomy of steering angle prediction has been presented in the paper for a better comprehension of the problem. We have also discussed open research problems at the end of the paper to help the researchers of this area to discover new research horizons.

Preventing COVID-19 Spread Using Information and Communication Technology

Without an effective vaccine, treatment, or therapy, the Coronavirus Disease 2019 (COVID-19) is spreading like fire and claiming lives. Countries began to adopt various strategies such as lockdown, mass testing, tracing, quarantine, sanitization, isolation, and treatment to contain COVID-19. However, it was soon realized that we need to take the help of powerful technologies to combat the spread of deadly COVID-19 until a vaccine or a drug is discovered. In this article, we discuss how the use of cutting edge technologies such as the Internet of Things (IoT), Big data, artificial intelligence (AI), unmanned aerial vehicles (UAVs)/drones, blockchain, robotics, autonomous ground vehicles, communication technologies in screening, testing, contact tracing, spread analysis, sanitization, and protocol enforcements can help prevent the COVID-19 spread.

A Framework for the Synergistic Integration of Fully Autonomous Ground Vehicles With Smart City

Most of the vehicle manufacturers aim to deploy level-5 fully autonomous ground vehicles (FAGVs) on city roads in 2021 by leveraging extensive existing knowledge about sensors, actuators, telematics and Artificial Intelligence (AI) gained from the level-3 and level-4 autonomy. FAGVs by executing non-trivial sequences of events with decimetre-level accuracy live in Smart City (SC) and their integration with all the SC components and domains using real-time data analytics is urgent to establish better swarm intelligent systems and a safer and optimised harmonious smart environment enabling cooperative FAGVs-SC automation systems. The challenges of urbanisation, if unmet urgently, would entail severe economic and environmental impacts. The integration of FAGVs with SC helps improve the sustainability of a city and the functional and efficient deployment of hand over wheels on robotized city roads with behaviour coordination. SC can enable the exploitation of the full potential of FAGVs with embedded centralised systems within SC with highly distributed systems in a concept of Automation of Everything (AoE). This article proposes a synergistic integrated FAGV-SC holistic framework - FAGVinSCF in which all the components of SC and FAGVs involving recent and impending technological advancements are moulded to make the transformation from today's driving society to future's next-generation driverless society smoother and truly make self-driving technology a harmonious part of our cities with sustainable urban development. Based on FAGVinSCF, a simulation platform is built both to model the varying penetration levels of FAGV into mixed traffic and to perform the optimal self-driving behaviours of FAGV swarms. The results show that FAGVinSCF improves the urban traffic flow significantly without huge changes to the traffic infrastructure. With this framework, the concept of Cooperative Intelligent Transportation Systems (C-ITS) is transformed into the concept of Automated ITS (A-ITS). Cities currently designed for cars can turn into cities developed for citizens using FAGVinSCF enabling more sustainable cities.

Trajectory tracking control of steer-by-wire autonomous ground vehicle considering the complete failure of vehicle steering motor

This paper investigates the trajectory tracking control problem of steer-by-wire autonomous ground vehicle considering the complete failure of vehicle steering motor. In order to achieve the trajectory tracking control of autonomous ground vehicle in presence of the complete failure of vehicle steering motor, the mechanical transmission mechanism of vehicle steering system is analyzed with the steering motor fault being considered, and the differential steering is used to actuate the turning of vehicle in case of emergency. A model predictive controller with the constraints of tire cornering angle and road adhesion is designed to compute the expected front-wheel steering angle for vehicle trajectory tracking control, and a nonsingular terminal sliding mode controller is proposed to track the expected front-wheel steering angle using differential-steering moment, in which a novel observer design method is presented to estimate the actual front-wheel steering angle. Two case studies are implemented in a CarSim-Simulink co-simulation platform, and the simulation results have verified that the proposed method have satisfactory trajectory tracking control effectiveness in presence of the complete failure of vehicle steering motor.

Vibration noise suppression approach based on random switching frequency control for permanent magnet motor

The vibration noise of the distributed drive autonomous ground vehicle (AGV) is mainly produced by the inverter-fed brushless DC permanent magnet in-wheel motor (PMIWM). It is necessary to reduce the vibration noise level of the PMIWM driven by pulse width modulation (PWM). A suppression approach of electromagnetic vibration noise for the PMIWM is investigated. First, the air gap magnetic field of the inverter-fed PMIWM was analyzed. The electric current harmonics and the unbalanced magnetic force (UMF) were investigated. The natural frequency and noise of the PMIWM were presented. Then, a flux-weaken approach was employed to maintain the robustness of the PMIWM when a sudden drop of DC bus voltage. The random switching frequency (RSF) PWM control method based on a two-state Markov chain is proposed to decrease the amplitude of the harmonics caused by the switching frequency and the multiple switching frequencies. The experimental results show that the RSFPWM can not only effectively reduce the vibration noise and the inverter losses, but also improve the robustness of the PMIWM control system under unpredictable uncertainties.

A reinforcement learning‐based approach for modeling and coverage of an unknown field using a team of autonomous ground vehicles

Precision maps are useful in agricultural farm management for providing farmers (and field researchers) with locational information. Having an environmental model that includes geo-referenced data would facilitate the deployment of multi-robot systems, that has emerged in precision agriculture. It further allows developing automated techniques and tools for map reconstruction and field coverage for farming purposes. In this study, a reinforcement learning-based method (and in particular dyna-Q+) is presented for a team of unmanned ground vehicles (UGVs) to cooperatively learn an unknown dynamic field (and in particular, an agricultural field). The problem we address here is to deploy UGVs to map plant rows, find obstacles, whose locations are not known a priori, and define regions of interest in the field (e.g., areas with high water stress). Once an environment model is built, the UGVs are then distributed to provide full coverage of plants and update the reconstructed map simultaneously. Simulation results are finally presented to demonstrate that the proposed method for simultaneous learning and planning can successfully learn a model of the field and monitor the coverage area.

Decentralised indoor smart camera mapping and hierarchical navigation for autonomous ground vehicles

In this work, the authors propose a novel decentralised coordination scheme for autonomous ground vehicles to enable map building and path planning with a network of smart overhead cameras. Decentralised indoor smart camera mapping and hierarchical navigation supports the automatic generation of waypoint graphs for each camera in an environment and allows path planning through the environment across multiple camera fields of view, or subviews. The proposed solution utilises the growing neural gas algorithm to learn the topology of unoccupied working space in each subview for maintaining a dynamic waypoint graph on each camera. The authors’ pathing solution leverages a modified version of the A* algorithm to compute paths in a decentralised and hierarchical fashion. Waypoint generation was simulated and analysed on a generated environment to ensure it is both effective and efficient, while path planning was simulated on various randomised hierarchical graphs to effectively compare the proposed Decentralised‐A* (D‐A*) algorithm against standard greedy search. The proposed method efficiently handles the cases where other robot navigation methods are otherwise weak and ineffective, while still providing avenues for further optimisation of resource overhead for both the smart camera network as well as the robots themselves.

Robust Set-Invariance Based Fuzzy Output Tracking Control for Vehicle Autonomous Driving Under Uncertain Lateral Forces and Steering Constraints

This paper is concerned with a new control method for path tracking of autonomous ground vehicles. We exploit the fuzzy model-based control framework to deal with the time-varying feature of the vehicle speed and the highly uncertain behaviors of the tire-road forces involved in the nonlinear vehicle dynamics. To avoid using costly vehicle sensors for feedback control while favoring the simplest control structure for real-time implementation, a new fuzzy static output feedback (SOF) scheme is proposed. In particular, though the robust set-invariance property and Lyapunov-based arguments, the physical constraints on the steering input saturation and the vehicle state can be taken into account in the control design to improve the driving safety and comfort. The theoretical development relies on the use of fuzzy Lyapunov functions and the non-parallel distributed compensation control concept to reduce the design conservatism. Exploiting some specific convexification techniques, the control design is reformulated as an optimization problem under linear matrix inequalities with a single line search, which are efficiently solved via semidefinite programming techniques. The proposed fuzzy path tracking controller is evaluated through various dynamic driving tests conducted with high-fidelity CarSim/Matlab co-simulations. Moreover, to emphasize the advantages of the new fuzzy SOF controller, a performance comparison with the CarSim driver model is also performed.

Vision-based robust control framework based on deep reinforcement learning applied to autonomous ground vehicles

Given the recent advances in computer vision, image processing and control systems, self-driving vehicles has been one of the most promising and challenging research topics nowadays. The design of vision-based robust controllers to keep an autonomous car in the center of the lane, despite uncertainties and disturbances, is still an ongoing challenge. This paper presents a hybrid control architecture that combines Deep Reinforcement Learning (DRL) and Robust Linear Quadratic Regulator (RLQR) for vision-based lateral control of an autonomous vehicle. Evolutionary estimation is used to model the vehicle uncertainties. For performance comparison, a DRL method and three other hybrid controllers are also evaluated. The inputs for each controller are real-time semantically segmented RGB camera images which serve as the basis to calculate continuous steering actions to keep the vehicle on the center of the lane with a constant velocity. Simulation results show that the proposed hybrid RLQR with evolutionary estimation of uncertainties architecture outperforms the other algorithms implemented. It presents lower tracking errors, smoother steering inputs, total collision avoidance and better generalization in new urban environments. Furthermore, it significantly decreases the required training time.

Super-twisting algorithm second-order sliding mode control for collision avoidance system based on active front steering and direct yaw moment control

Active collision avoidance system has received more and more attraction, which has the capability to avoid potential accidents and reduce driver burden. This paper proposes an active collision avoidance system which consists of a path planner and a coordinated lateral controller. In the path planner, cubic B-spline is developed to obtain collision-free trajectories to bypass the obstacle by steering. Based on this, a coordinated lateral dynamic control of autonomous ground vehicles is presented to improve the accuracy and robustness of path following and simultaneously ensure vehicle stability via active front steering and direct yaw moment control. Then, second-order sliding mode control, based on super-twisting algorithm, is applied to reduce lateral offset and heading angle deviation as much as possible and avoid chattering phenomenon of tradition sliding mode control. Meanwhile, a new form of sliding mode control based on improved reaching law is devoted to forcing the vehicle state sideslip angle and yaw rate to stability envelope with less chattering in the case of low road friction coefficient. Eventually, the effectiveness and robustness of active collision avoidance system against external disturbance and parametric uncertainties are confirmed through different test cases in the MATLAB/Simulink simulation platform.

SYSTEMATIC EVALUATION AND CHARACTERIZATION OF 3D SOLID STATE LIDAR SENSORS FOR AUTONOMOUS GROUND VEHICLES

Abstract. 3D LiDAR sensors play an important part in several autonomous navigation and perception systems with the technology evolving rapidly over time. This work presents the preliminary evaluation results of a 3D solid state LiDAR sensor. Different aspects of this new type of sensor are studied and their data are characterized for their effective utilization for object detection for the application of Autonomous Ground Vehicles (AGV). The paper provides a set of evaluations to analyze the characterizations and performances of such LiDAR sensors. After characterization of the sensor, the performance is also evaluated in real environment with the sensors mounted on top of a vehicle and used to detect and classify different objects using a state-of-the-art Super-Voxel based method. The 3D point cloud obtained from the sensor is classified into three main object classes “Building”, “Ground” and “Obstacles”. The results evaluated on real data, clearly demonstrate the applicability and suitability of the sensor for such type of applications.

Assessing Impact of Understory Vegetation Density on Solid Obstacle Detection for Off-Road Autonomous Ground Vehicles

Abstract In autonomous driving systems, advanced sensing technologies (such as Light Detection and Ranging (LIDAR) devices and cameras) can capture high volume of data for real-time traversability analysis. Off-road autonomy is more challenging than other autonomous applications due to the highly unstructured environment with various types of vegetation. The understory with unknown density can create extremely challenging scenarios (such as negative obstacles masked by dense vegetation) by concealing potential obstacles in the terrain, leading to severe vehicle damage, significant financial loss, and even operator injury or death. This paper investigates the impact of understory vegetation density on obstacle detection in off-road traversability analysis. By leveraging a physics-based autonomous driving simulator, a machine learning–based framework is proposed for obstacle detection based on point cloud data captured by LIDAR. It is observed that the increase in the density of understory vegetation adversely affects the classification performance in correctly detecting solid obstacles. With the cumulative approach used in this paper, however, sensitivity results for different density levels converge as the vehicles incorporates more time frame data into the classification algorithm.