Control Of Autonomous Vehicles Research Articles

Event-Triggered Quantitative Prescribed Performance Neural Adaptive Control for Autonomous Underwater Vehicles

Event-Triggered Quantitative Prescribed Performance Neural Adaptive Control for Autonomous Underwater Vehicles

A novel adaptive trajectory tracking control for autonomous vehicles based on state expansion

A novel adaptive trajectory tracking control for autonomous vehicles based on state expansion

Research on cooperative control of vehicle path tracking and lateral stability

Abstract This study presents a novel trajectory tracking control approach integrating steering angle and wheel torque output. Unlike conventional hierarchical methods, it considers front wheel steering angle and direct swing torque control simultaneously. Utilizing Model Predictive Control for autonomous vehicles, the system employs path tracking deviation as the state variable, front wheel steering angle, and overall additional yaw moment as control inputs. The coordinated control strategy enhances vehicle stability and contributes to heading angle changes, improving trajectory tracking accuracy. Simulation results in Matlab/Simulink demonstrate the validity and superiority of the proposed method in trajectory tracking and vehicle stability.

Perception simplex: Verifiable collision avoidance in autonomous vehicles amidst obstacle detection faults

AbstractAdvances in deep learning have revolutionized cyber‐physical applications, including the development of autonomous vehicles. However, real‐world collisions involving autonomous control of vehicles have raised significant safety concerns regarding the use of deep neural networks (DNNs) in safety‐critical tasks, particularly perception. The inherent unverifiability of DNNs poses a key challenge in ensuring their safe and reliable operation. In this work, we propose perception simplex (), a fault‐tolerant application architecture designed for obstacle detection and collision avoidance. We analyse an existing LiDAR‐based classical obstacle detection algorithm to establish strict bounds on its capabilities and limitations. Such analysis and verification have not been possible for deep learning‐based perception systems yet. By employing verifiable obstacle detection algorithms, identifies obstacle existence detection faults in the output of unverifiable DNN‐based object detectors. When faults with potential collision risks are detected, appropriate corrective actions are initiated. Through extensive analysis and software‐in‐the‐loop simulations, we demonstrate that provides deterministic fault tolerance against obstacle existence detection faults, establishing a robust safety guarantee.

Robust autonomous vehicle control by leveraging multi-stage MPC and quantized CNN in HIL Framework

This study presents a novel path planning and path-following system designed explicitly for real-world driving conditions in autonomous vehicles. The pipeline considers uncertainties in the dynamic parameters of the vehicle, which are common in real-world scenarios. To address these uncertainties, a robust multi-stage model predictive control approach is developed to generate optimal control outputs. A customized convolutional neural network, Nano-FastSCNN, quantized to half-precision, is employed for accurately segmenting left and right lane boundaries. To ensure that our visual planner is robust against adverse weather conditions, we have rigorously evaluated its generalization capabilities under challenging weather types. The centerline of these lanes is then selected as the desired path and transformed into the camera reference frame by leveraging the camera projection matrix. The controller utilizes this reference trajectory to guide the vehicle along the desired trajectory. To demonstrate real-time processing capabilities, the entire pipeline, including the vision system module and controller, is implemented in a hardware-in-the-loop setup utilizing three NVIDIA Jetson devices to demonstrate the scalability of our architecture. Comprehensive evaluations demonstrate that the pipeline achieves an inference speed of 15.6 FPS, even on the less powerful NVIDIA Jetson Nano. Moreover, the system successfully alleviates the detrimental effects of uncertainties and reliably follows the desired path.

Remote path-following control for a holonomic Mecanum-wheeled robot in a resource-efficient networked control system

This paper introduces a novel resource-efficient control structure for remote path-following control of autonomous vehicles based on a comprehensive combination of Kalman filtering, non-uniform dual-rate sampling, periodic event-triggered communication, and prediction-based and packet-based control techniques. An essential component of the control solution is a non-uniform dual-rate extended Kalman filter (NUDREKF), which includes an h-step ahead prediction stage. The prediction error of the NUDREKF is ensured to be exponentially mean-square bounded. The algorithmic implementation of the filter is straightforward and triggered by periodic event conditions. The main goal of the approach is to achieve efficient usage of resources in a wireless networked control system (WNCS), while maintaining satisfactory path-following behavior for the vehicle (a holonomic Mecanum-wheeled robot). The proposal is additionally capable of coping with typical drawbacks of WNCS such as time-varying delays, and packet dropouts and disorder. A Simscape Multibody simulation application reveals reductions of up to 93% in resource usage compared to a nominal time-triggered control solution. The simulation results are experimentally validated in the holonomic Mecanum-wheeled robotic platform.

Research on a Personalized Decision Control Algorithm for Autonomous Vehicles Based on the Reinforcement Learning from Human Feedback Strategy

To address the shortcomings of previous autonomous decision models, which often overlook the personalized features of users, this paper proposes a personalized decision control algorithm for autonomous vehicles based on RLHF (reinforcement learning from human feedback). The algorithm combines two reinforcement learning approaches, DDPG (Deep Deterministic Policy Gradient) and PPO (proximal policy optimization), and divides the control scheme into three phases including pre-training, human evaluation, and parameter optimization. During the pre-training phase, an agent is trained using the DDPG algorithm. In the human evaluation phase, different trajectories generated by the DDPG-trained agent are scored by individuals with different styles, and the respective reward models are trained based on the trajectories. In the parameter optimization phase, the network parameters are updated using the PPO algorithm and the reward values given by the reward model to achieve personalized autonomous vehicle control. To validate the control algorithm designed in this paper, a simulation scenario was built using CARLA_0.9.13 software. The results demonstrate that the proposed algorithm can provide personalized decision control solutions for different styles of people, satisfying human needs while ensuring safety.

Using the Buckingham π Theorem for Multi-System Transfer Learning: A Case-Study with 3 Vehicles Sharing a Database

Many advanced driver assistance schemes or autonomous vehicle controllers are based on a motion model of the vehicle behavior, i.e., a function predicting how the vehicle will react to a given control input. Data-driven models, based on experimental or simulated data, are very useful, especially for vehicles difficult to model analytically, for instance, ground vehicles for which the ground-tire interaction is hard to model from first principles. However, learning schemes are limited by the difficulty of collecting large amounts of experimental data or having to rely on high-fidelity simulations. This paper explores the potential of an approach that uses dimensionless numbers based on Buckingham’s π theorem to improve the efficiency of data for learning models, with the goal of facilitating knowledge sharing between similar systems. A case study using car-like vehicles compares traditional and dimensionless models on simulated and experimental data to validate the benefits of the new dimensionless learning approach. Preliminary results from the case study presented show that this new dimensionless approach could accelerate the learning rate and improve the accuracy of the model prediction when transferring the learned model between various similar vehicles. Prediction accuracy improvements with the dimensionless scheme when using a shared database, that is, predicting the motion of a vehicle based on data from various different vehicles was found to be 480% more accurate for predicting a simple no-slip maneuver based on simulated data and 11% more accurate to predict a highly dynamic braking maneuver based on experimental data. A modified physics-informed learning scheme with hand-crafted dimensionless features was also shown to increase the improvement to precision gains of 917% and 28% respectively. A comparative study also shows that using Buckingham’s π theorem is a much more effective preprocessing step for this task than principal component analysis (PCA) or simply normalizing the data. These results show that the use of dimensionless variables is a promising tool to help in the task of learning a more generalizable motion model for vehicles, and hence potentially taking advantage of the data generated by fleets of vehicles on the road even though they are not identical.

Machine Learning-Based Control of Autonomous Vehicles for Solar Panel Cleaning Systems in Agricultural Solar Farms

This paper presents a machine learning (ML)-based approach for the intelligent control of Autonomous Vehicles (AVs) utilized in solar panel cleaning systems, aiming to mitigate challenges arising from uncertainties, disturbances, and dynamic environments. Solar panels, predominantly situated in dedicated lands for solar energy production (e.g., agricultural solar farms), are susceptible to dust and debris accumulation, leading to diminished energy absorption. Instead of labor-intensive manual cleaning, robotic cleaners offer a viable solution. AVs equipped to transport and precisely position these cleaning robots are indispensable for the efficient navigation among solar panel arrays. However, environmental obstacles (e.g., rough terrain), variations in solar panel installation (e.g., height disparities, different angles), and uncertainties (e.g., AV and environmental modeling) may degrade the performance of traditional controllers. In this study, a biologically inspired method based on Brain Emotional Learning (BEL) is developed to tackle the aforementioned challenges. The developed controller is implemented numerically using MATLAB-SIMULINK. The paper concludes with a comparative analysis of the AVs’ performance using both PID and developed controllers across various scenarios, highlighting the efficacy and advantages of the intelligent control approach for AVs deployed in solar panel cleaning systems within agricultural solar farms. Simulation results demonstrate the superior performance of the ML-based controller, showcasing significant improvements over the PID controller.

Personalized driving behavior oriented autonomous vehicle control for typical traffic situations

autonomous driving systems not only provide services for human drivers, but also need to consider the personalized driving requirements of human beings. In current road traffic environments, the driving behaviors of drivers differ significantly. This suggests that the various needs of different drivers cannot be met by a single behavior mode in an autonomous driving decision-making system. This paper looks at the personalized characteristics of various drivers and considers the implications of their differences. First, a Proportional Integral Differential (PID) feedback channel is introduced in a traditional Model Predictive Control (MPC) to improve the performance of the controller, and comprehensively considering the collision risk, motion hysteresis and rule constraints, referring to the MPC idea, a collision avoidance method based on Q-ABSAS optimization is proposed. Then based on the Chance Constrained Programming, the control constraint is combined with driver personalization to reflect a variety of driving personality characteristics. Finally, the proposed method is tested using Hardware-in-the-loop (HIL) experiments. The experiment results demonstrate that the proposed method can successfully implement vehicle tracking control and make the vehicle's state of movement match the driver's expectations, which can increase driver comfort and driving safety.

Adaptive path tracking control for autonomous vehicles with sideslip effects and unknown input delays

This paper presents an adaptive path-tracking control of autonomous ground vehicles with sideslip effects and time-varying input delays. To handle the problem of unknown time-varying input delays, the vehicle kinematic model is transformed into a novel chain model, and a delay-based auxiliary integral term is introduced for the estimation of the unknown terms caused by time-varying delays. To avoid modeling errors caused by existing linear approximation methods, a new adaptive path-tracking control strategy is proposed, such that the optimal tracking performance is achieved by appropriately adjusting certain design parameters. In addition, the inherent complexity explosion issue is avoided with the aid of a boundary estimation strategy. All signals of the closed-loop systems are guaranteed to be bounded. Simulation results illustrate the effectiveness of the method proposed.

UAV formation control using enhanced behavior mechanism and artificial potential field

Inspired by formation flight of pigeon flock, this paper proposes a enhanced method of autonomous formation control of multiple Unmanned Aerial Vehicles (UAVs) that can maintain high symmetry based on pigeon flock behavior mechanism. Addressing the instability of formation in the original method, the follow improvements have been made. Firstly, improve leadership of top three UAVs, Secondly, modify artificial potential field strategies for top two followers. Finally, through a series of simulation experiments, it is verified that the UAVs can form the expected formation under the autonomous formation control, and can maintain the formation under the complex motion of leader UAV.

PID-like IT2FLC-Based Autonomous Vehicle Control in Urban Areas

PID-like IT2FLC-Based Autonomous Vehicle Control in Urban Areas

Event‐triggered obstacle avoidance control for autonomous surface vehicles with actuator faults

AbstractThe article addresses the event‐triggered obstacle avoidance control problem for autonomous surface vehicles subject to actuator faults. In order to tackle the challenges presented by unknown actuator faults, internal model uncertainties, and external disturbances, we propose a fault‐tolerant control scheme that relies on an extended state observer to estimate the unknown parameters. In addition, considering the occurrence of actuator faults, a novel event‐triggered mechanism is developed to reduce wear of actuator and ensures the performance of system. On this basis, we employ backstepping technique and an improved artificial potential function methods to develop an event‐triggered obstacle avoidance control scheme with the capability of fault tolerance. This proposed control strategy ensures the uniform ultimate boundedness of tracking errors. In contrast to existing results, the presented control strategy simultaneously holds the performance of obstacle avoidance and fault tolerance and reduce the update frequency of actuators. The effectiveness of the provided control scheme has been confirmed through simulation.

Robust optimal tracking control of multiple autonomous underwater vehicles subject to uncertain disturbances

AbstractThis paper considers the problem of robust optimal tracking control of multiple autonomous underwater Vehicles (AUVs) subject to uncertain external disturbances. First, the Takagi‐Sugeno (T‐S) fuzzy based technique is utilized to convert the high‐order nonlinear multi‐AUV system into a series of linearized subsystems. Second, a novel fully distributed sliding mode control (FDSMC) strategy is proposed to attenuate the disturbances. Meanwhile, the leader‐following consensus and the nearly optimization of the energy‐cost function for the multi‐AUV system can be achieved simultaneously through the designed optimal nominal control protocol. Moreover, the proposed control strategy has more mild constraints on the communication topologies. Finally, the effectiveness of the proposed FDSMC strategy is verified by numerical simulation studies.

Accurate data-driven sliding mode parking control for autonomous ground vehicles with efficient trajectory planning in dynamic industrial scenarios

Accurate data-driven sliding mode parking control for autonomous ground vehicles with efficient trajectory planning in dynamic industrial scenarios

A Model-Based Adaptive Control of Turning Maneuver for Catamaran Autonomous Surface Vessel

The effective computer control of airborne, waterborne, and ground autonomous vehicles has become one of the highest priorities in the area of cyber-physical systems, Industry 4.0 in particular, and the world economy in general. Despite extensive research on Unmanned Aerial Vehicles (UAVs), the study of Autonomous Surface Vessels (ASVs) has been more than ten times less intense. As an attempt to fill a gap in that field, this article discusses the control mathematics of a realistic ASV nonlinear model of a real autonomous electric catamaran “Nymo”, which was designed at Tallinn University of Technology (TalTech). More technically, the article offers a novel adaptive control system that is based on knowledge of the main parameters of ASV and is specially designed for a Simulink/MATLAB environment. The article also enables adjusting variables like transition time and heading angle overshoot value. The control of the desired tracking is represented in such a maneuver as turning the catamaran at different angles. The designed control system has shown good quality in terms of accuracy in tracking the desired heading angles.

Hand Gesture Controlled Smart Car Using Image Recognition

A hand gesture-controlled smart automobile is the newest initiative in the field of human-computer interaction, and it represents the evolution toward more natural and in- tuitive user interfaces. This paper describes how OpenCV and Google’s MediaPipe are intricately coordinated to produce a control strategy that is both agile and responsive. The deployed technology translates dynamic vehicle motion commands from complex human hand gestures using advanced image recognition algorithms. This is the pinnacle of interactive technology meeting real-world locomotion needs: state-of-the-art computer vision and machine learning combined. The suggested system not only demonstrates the viability of non-contact user input for complex tasks like driving, but it also sets the way for future research in the field of autonomous vehicle guidance and control systems. This study highlights how gesture-based interfaces have the power to completely change the way people interact with cars, paving the way for more flexible and human-centered navigation systems.

Observer-based prescribed performance path-following control for autonomous ground vehicles via error shifting method

Observer-based prescribed performance path-following control for autonomous ground vehicles via error shifting method

Neural network-based practical fixed-time nonsingular sliding mode tracking control of autonomous surface vehicles under actuator saturation

The article considers the fixed-time tracking control of autonomous surface vehicles (ASVs) subject to actuator saturation. A novel sliding variable with state-dependent exponents is first constructed to alleviate chattering and overcome the singularity. By using the proposed fixed-time sliding variable and the neural network (NN), an NN-based fixed-time sliding mode control (SMC) scheme is developed. Besides, an auxiliary system is utilized to solve the input saturation constraint. Unlike some existing results for ASVs with input saturation, the designed fixed-time SMC law does not require assumptions on a priori upper bound of the lumped disturbances. Through Lyapunov analysis, it is proven that the tracking error can converge to a bounded region in a fixed time. Simulation examples are given to show the validity of the proposed control scheme.