Autonomous, or self-driving, cars are emerging as the solution to several problems primarily caused by humans on roads, such as accidents and traffic congestion. However, those benefits come with great challenges in the verification and validation (V&V) for safety assessment. In fact, due to the possibly unpredictable nature of Artificial Intelligence (AI), its use in autonomous cars creates concerns that need to be addressed using appropriate V&V processes that can address trustworthy AI and safe autonomy. In this study, the relevant research literature in recent years has been systematically reviewed and classified in order to investigate the state-of-the-art in the software V&V of autonomous cars. By appropriate criteria, a subset of primary studies has been selected for more in-depth analysis. The first part of the review addresses certification issues against reference standards, challenges in assessing machine learning, as well as general V&V methodologies. The second part investigates more specific approaches, including simulation environments and mutation testing, corner cases and adversarial examples, fault injection, software safety cages, techniques for cyber-physical systems, and formal methods. Relevant approaches and related tools have been discussed and compared in order to highlight open issues and opportunities.

/pdf/software-verification-and-validation-of-safe-autonomous-cars-5n59tyoh6v.pdf

Software Verification and Validation of Safe Autonomous Cars: A Systematic Literature Review

Road traffic signals and traffic infrastructure have a significant impact on the behaviour of highly automated or autonomous vehicles. However, the increase in automation does not always mean an advantage. Generally, highly automated vehicles strictly follow the traffic rules resulting in near-accident situations, although their goal is to avoid and reduce them. Malfunctions of the automated functions might cause surprising interventions while the vehicle is in motion, drivers cannot react in time and well. This paper highlights the potential danger and uncertainty of highly automated or autonomous vehicles in the context of the current conventional traffic infrastructure system. In the future, special consideration shall be given to the vehicle industry and traffic regulation makers on how the infrastructure should be adapted to automated vehicle functions to have a seamless shift towards automated driving. The paper sums up many problematic situations with two critical problems with sensitivity analysis. The first situation: the speed assist system (based on speed limit sign recognition) conflicts with the traffic infrastructure. The second situation is shown: the ACC (Adaptive Cruise Control) and LKA (Lane Assist) contradicts with the traffic infrastructure. These critical situations were investigated by using an high-fidelity automotive simulation software as proof of concept and were examined by accident reconstruction analysis software.

/pdf/conflicts-of-automated-driving-with-conventional-traffic-4ntgmc4atc.pdf

Conflicts of Automated Driving With Conventional Traffic Infrastructure

In this contribution, we suggest two proposals to achieve fast, real-time lane-keeping control for Autonomous Ground Vehicles (AGVs). The goal of lane-keeping is to orient and keep the vehicle within a given reference path using the front wheel steering angle as the control action for a specific longitudinal velocity. While nonlinear models can describe the lateral dynamics of the vehicle in an accurate manner, they might lead to difficulties when computing some control laws such as Model Predictive Control (MPC) in real time. Therefore, our first proposal is to use a Linear Parameter Varying (LPV) model to describe the AGV’s lateral dynamics, as a trade-off between computational complexity and model accuracy. Additionally, AGV sensors typically work at different measurement acquisition frequencies so that Kalman Filters (KFs) are usually needed for sensor fusion. Our second proposal is to use a Dual-Rate Extended Kalman Filter (DREFKF) to alleviate the cost of updating the internal state of the filter. To check the validity of our proposals, an LPV model-based control strategy is compared in simulations over a circuit path to another reduced computational complexity control strategy, the Inverse Kinematic Bicycle model (IKIBI), in the presence of process and measurement Gaussian noise. The LPV-MPC controller is shown to provide a more accurate lane-keeping behavior than an IKIBI control strategy. Finally, it is seen that Dual-Rate Extended Kalman Filters (DREKFs) constitute an interesting tool for providing fast vehicle state estimation in an AGV lane-keeping application.

https://www.mdpi.com/1424-8220/21/4/1531/pdf

Autonomous Ground Vehicle Lane-Keeping LPV Model-Based Control: Dual-Rate State Estimation and Comparison of Different Real-Time Control Strategies.

. Air pollution, energy consumption, and human safety
issues have aroused people's concern around the world. This phenomenon could
be significantly alleviated with the development of automatic driving
techniques, artificial intelligence, and computer science. Autonomous
vehicles can be generally modularized as environment perception, path
planning, and trajectory tracking. Trajectory tracking is a fundamental part
of autonomous vehicles which controls the autonomous vehicles effectively
and stably to track the reference trajectory that is predetermined by the
path planning module. In this paper, a review of the state-of-the-art trajectory
tracking of autonomous vehicles is presented. Both the trajectory tracking
methods and the most commonly used trajectory tracking controllers of
autonomous vehicles, besides state-of-art research studies of these controllers,
are described.

/pdf/review-article-state-of-the-art-trajectory-tracking-of-5ftnaf8utb.pdf

Review article: State-of-the-art trajectory tracking of autonomous vehicles

Professor drivers, including racing drivers, can drive cars to achieve drift motions by taking control of the steering angle in high tire slip ratios, which provides a way to improve the driving safety of autonomous vehicles. The existing studies can be divided into two kinds based on analysis methods, and the theory-based is chosen in this study. Because the recent theory based is most applied for planar models with neglect of the rollover accident risk, the nonlinear vehicle model is established by considering longitudinal, lateral, roll, and yaw motions and rolling safety with the nonlinear tire model UniTire. The drift motion mechanism is analyzed in steady and transient states to obtain drift motion conditions, including the velocity limitation and the relationship between sideslip angle and yaw rate, and vehicle main status parameters including the velocity, side-slip angle and yaw rate in drift conditions. The state-feedback controller is designed based on robust theory and LMI (linear matrix inequation) with uncertain disturbances to realize circle motions in drift conditions. The designed controller in simulations realizes drift circle motions aiming at analyzed status target values by matching the front-wheel steering angle with saturated tire forces, which satisfies the Lyapunov stability with robustness. Robust control in drift conditions solves the problem of how to control vehicles to perform drift motions with uncertain disturbances and improves the driving safety of autonomous vehicles.

Robust Control with Uncertain Disturbances for Vehicle Drift Motions

The driver assistance systems becomes more common in today’s vehicles, and in the future highly automated vehicles and large number of autonomous vehicles will appear in roads. In the beginning these systems only support the human driver, but later they will able to control the vehicle driving. Hence vehicles have to plan the path, and follow it accurately and safely. Furthermore it has to provide convenience for the passengers. To find a better trajectory following controller first it is necessary to test it in simulation environment, then to test it in real environment in a vehicle.The subject of this article is an implementation of a lateral dynamic bicycle model, and an implementation of different controllers. Besides it will contain the testing of the controllers. If the controllers work with sufficient precision in the simulation, it will be tested in a real vehicle. The essence of the test is the speed limit, where the simulation and the real test starting to deviate. In the end of the article we compare the performance of the different controllers and evaluate which is the most appropriate. Besides the difference will also be evaluated between the simulated results and the results on the real vehicle. This comparison reveal which is the speed limit, as long as the values calculated on the lateral dynamic bicycle model approximate the actual results in a simulation environment with sufficient precision.

Comparison of path following controllers for autonomous vehicles

In recent years vehicles with high level of automation are becoming more popular which is expected to continue in the future. Advanced driving assistance systems are increasingly taking control of the vehicle, starting with the support of the driving task. Automated or highly automated vehicles are expected to follow a planned road safely. In this paper, the authors aim to improve the accuracy of path following by developing a new control strategy. The controller includes both pure pursuit and Stanley methods, the operation of the controller is based on the geometry of the vehicle and the path; the key is the proper weighting between two controllers. The performance of the combined path following controller was measured on a demonstration vehicle.

Path following controller for autonomous vehicles

Electronic vehicle dynamics systems are expected to evolve in the future as more and more automobile manufacturers mark fully automated vehicles as their main path of development. State-of-the-art electronic stability control programs aim to limit the vehicle motion within the stable region of the vehicle dynamics, thereby preventing drifting. On the contrary, in this paper, the authors suggest its use as an optimal cornering technique in emergency situations and on certain road conditions. Achieving the automated initiation and stabilization of vehicle drift motion (also known as powerslide) on varying road surfaces means a high level of controllability over the vehicle. This article proposes a novel approach to realize automated vehicle drifting in multiple operation points on different road surfaces. A three-state nonlinear vehicle and tire model was selected for control-oriented purposes. Model predictive control (MPC) was chosen with an online updating strategy to initiate and maintain the drift even in changing conditions. Parameter identification was conducted on a test vehicle. Equilibrium analysis was a key tool to identify steady-state drift states, and successive linearization was used as an updating strategy. The authors show that the proposed controller is capable of initiating and maintaining steady-state drifting. In the first test scenario, the reaching of a single drifting equilibrium point with −27.5° sideslip angle and 10 m/s longitudinal speed is presented, which resulted in −20° roadwheel angle. In the second demonstration, the setpoints were altered across three different operating points with sideslip angles ranging from −27.5° to −35°. In the third test case, a wet to dry road transition is presented with 0.8 and 0.95 road grip values, respectively.

Model Predictive Controller Design for Vehicle Motion Control at Handling Limits in Multiple Equilibria on Varying Road Surfaces

The presence of highly automated vehicles and driver assistance systems is expected to grow in the future. In recent years, these systems offer assistance for the driver but, in the future, the human driver will be increasingly replaced by automated vehicle functions. In order to achieve this, the vehicle must be able to plan and safely follow a trajectory. In this paper, a MIMO linear quadratic regulator was designed for stabilizing the vehicle during steady-state drifting. The research can be utilized in different fields of applications. Namely, the driving capabilities of an automated vehicle must be at least as good as a human driver. Hence, these systems must be able to drive the car at friction limits. Furthermore, a better understanding of the dynamics of drifting and the equilibrium conditions can be used in motorsports. A three-state bicycle model has been implemented, supplemented with two different tire models. A brush tire model was used at front wheels. A combined slip brush tire model was used at rear wheels, considering lateral-longitudinal force coupling. During drift, the rear tires are saturated, the front tires are not. The complete vehicle model containing the tire models described by eight equations, the system of equations has been solved numerically for equilibrium points. The nonlinear vehicle model was Jacobian linearized in a parametric form. Based on the results, a MIMO LQ system was designed. The control inputs include the steering angle at front wheels and longitudinal drive force at rear wheels. Finally, the controller performance was demonstrated using the nonlinear model in MATLAB/Simulink environment. The proposed controller found able to regulate steady-state drifting with predefined sideslip angles and longitudinal velocities.

MIMO Controller Design for Stabilizing Vehicle Drifting

In the future, the appearance of highly automated vehicles is expected, the driver assistance systems will become more common in the vehicles. Initially, these systems are only supported the driver, but increasingly take over the control of the entire vehicle. In this practice, the vehicle must plan the route and follow it safely while providing a comfortable journey for passengers. In order to develop the path following controller, it is first tested in a simulation environment and then on a real vehicle. In a previous paper, authors have implemented different path following controllers that were tested on a demonstration vehicle after testing in the simulation environment. In the simulation environment, the controllers were tested on a dynamic bicycle model, which model describes the lateral dynamics of the vehicle. The simulation and measurement results showed a difference. The subject of this paper is to find out the reason of the difference that was identified in the dynamic behavior of the steering system, including the effect of the tires. Based on these, the vehicle model was adjusted. Measurements were made to identify the dynamics of the steering system, and the tires. Analyzing the results of the measurements, the dynamic model of the steering system was described, the vehicle model has been expanded on this basis. Finally, the path following simulation with the improved model was calculated, which was compared to the real vehicle measurement and the results showed good correlation.

Adam Domina

Papers

Comparison of path following controllers for autonomous vehicles

Path following controller for autonomous vehicles

Model Predictive Controller Design for Vehicle Motion Control at Handling Limits in Multiple Equilibria on Varying Road Surfaces

MIMO Controller Design for Stabilizing Vehicle Drifting

Modelling the dynamic behavior of the steering system for low speed autonomous path tracking