No description available.

Thesis (Ph. D.)--Virginia Polytechnic Institute and State University, 1996. / Vita. Abstract. Includes bibliographical references (leaves 166-171). Also available via the Internet.

For decades vehicle control has been extensively studied to investigate and improve vehicle stability and performance. Such controllers are designed to improve driving safety while the driver is still in control of the vehicle. It is known that human drivers are capable to learn and adapt to their built-in vehicle controller in order to improve their control actions based on their past driving experiences with the same vehicle controller. Although the learning curve varies for different human drivers, it results in a more constructive cooperation between the human driver and the computer-based vehicle controller, leading to globally optimal vehicle stability. The main intent of this research is to develop a novel cooperative interaction model between the human driver and vehicle controller in order to obtain globally optimal vehicle steering and lateral control. Considering the vehicle driver-controller interactions as a common two-player game problem where both players attempt to improve their payoffs, i.e., minimize their objective functions, the Game Theory approach is applied to obtain the optimal driver's steering inputs and controller's corrective yaw moment. Extending this interaction model to include more realistic scenarios, the model is discretized and a road preview model is added to account for the driver's preview-time characteristic. Also, a robust interaction model is developed to stabilize the vehicle performance while taking bounded uncertainty effects in driver's steering behavior into consideration using the Integral Sliding Mode control methodology. For evaluation purposes, a nonlinear vehicle dynamics model is developed that captures nonlinear tire characteristics and includes driver steering controllability and vehicle speed control systems such as cruise control, differential braking, and anti-lock braking systems. A graphical user interface (GUI) is developed in MATLAB to ease the use of the vehicle model and hopefully encourage its widespread application in the future. Simulation results indicate that the proposed cooperative interaction model, which is the end-product of human driver's and vehicle controller's mutual understanding of each other's objective and performance quality, results in more optimal and stable vehicle performance in lateral and yaw motions compared to the existing LQR controllers that tend to independently optimize the driver and vehicle controller inputs. / Ph. D.

The gradual decline in global oil reserves and the presence of ever so stringent emissions rules around the world have created an urgent need for the production of automobiles with improved fuel economy. HEVs (hybrid electric vehicles) have proved a viable option to guaranteeing improved fuel economy and reduced emissions. The fuel consumption benefits which can be realised when utilising HEV architecture are dependent on how much braking energy is regenerated, and how well the regenerated energy is utilised. The challenge in developing a real-time HEV control strategy lies in the satisfaction of often conflicting control constraints involving fuel consumption, emissions and driveability without over-depleting the battery state of charge at the end of the defined driving cycle. Reviewed literature indicates some research gaps and hence exploitable study areas for which this thesis intends to address. For example, despite the research advances made, HEV energy management is still lacking in several key areas: optimisation of braking energy regeneration; real-time sub-optimal control of HEV for robustness, charge sustenance and fuel reduction; and real-time vehicle speed control. Consequently, this thesis aims to primarily develop novel real-time near-optimal control strategies for a parallel HEV, with a view to achieving robustness, fuel savings and charge sustenance simultaneously, under various levels of obtainable driving information (no route preview information, partial route preview information). Using a validated HEV dynamic simulation model, the following novel formulations are proposed in this thesis and subsequently evaluated in real time: 1. A simple grouping system useful for classifying standard and real-world driving cycles on the basis of aggressivity and road type. 2. A simple and effective near-optimal heuristic control strategy with no access to route preview information. 3. A dynamic programming-inspired real-time near-optimal control strategy with no access to route preview information. 4. An ECMS (Equivalent Consumption Minimisation Strategy) inspired real-time near-optimal control strategy with no access to route preview information. 5. An ECMS-inspired real-time near-optimal control strategy with partial access to route preview information. 6. A dynamic programming based route-optimal vehicle speed control strategy which accounts for real-time dynamic effects like engine braking, while solving an optimisation problem involving the maximisation of fuel savings with little or no penalty to trip time. 7. A real-time vehicle speed control approach, which is based on smoothing the speed trajectory of the lead vehicle, consequently reducing the acceleration and deceleration events that the intelligent vehicle (follower vehicle) will undergo. This smoothing effect translates into reduced fuel consumption, which tends to increase with increasing traffic preview window. Among other studies performed in this thesis, the fuel savings potential of the proposed near-optimal controllers was investigated in real time over standard driving cycles and real-world driving profiles. Results from these analyses show that, over standard driving cycles, properly formulated near-optimal real-time controllers are able to achieve a fuel savings potential within 0.03% to 3.71% of the global optimal performance, without requiring any access to route preview information. It was also shown that as much as 2.44% extra fuel savings could be achieved over a driving route, through the incorporation of route preview information into a real-time controller. Investigations were also made into the real-time fuel savings that could be realised over a driving route, through vehicle speed control. Results from these analyses show that, compared to an HEV technology which comes at a bigger cost, far higher fuel savings, as much as 45.96%, could be achieved through a simple real-time vehicle speed control approach.

Real-time path generation, including collision checks, is vital in critical driving scenarios such as navigating non-signalized intersections. These intersections lack organized traffic flow, which raises the risk of accidents. Rapidly Exploring Random Trees (RRT) is a widely adopted algorithm in robotics for motion planning due to its simplicity and probabilistic completeness. Over the years, researchers have made modifications to the basic RRT algorithm to improve its performance in dynamic environments, making it a favored planning algorithm for autonomous driving. Among these variants, probabilistic RRT (pRRT) demonstrates promising capabilities for efficient online replanning. The first part of the thesis thoroughly studies the pRRT algorithm and compares its performance to the standard RRT and RRT* algorithms through Python simulations. The pRRT algorithm outperformed the RRT and RRT* algorithms in terms of success rate and time to find a safe trajectory. The algorithm was implemented experimentally on scaled cars for the validation of its feasibility. The experimental results show good sim-to-real transfer for this algorithm. The second part of the thesis proposes a novel algorithm for path planning. The algorithm outperforms the standard RRT and pRRT techniques in terms of optimality and conformance to human instincts. The generated paths are much smoother and easier for the controller to track. The AV implementation combines the probabilistic RRT with the RRT-Connect algorithm to mitigate the problem of parameter tuning of the standard pRRT algorithm. The idea is to generate intermediate critical points around the obstacles to grow multiple trees between these points, which are then eventually connected if a safe trajectory is found. The algorithm was tested in simulation and showed comparatively better performance in handling obstacles. / Master of Science / Due to uncontrolled traffic flow, non-signalized intersections are critical for autonomous driving. Motion planning is responsible for the vehicle's decision-making and generating actions based on its surroundings. Rapidly Exploring Random Trees (RRT) is one of the most widely used algorithms for motion planning in robotics due to its simplicity and a guarantee of finding a collision-free path if it exists. Due to the randomness of the algorithm, the time to find a collision-free path increases rapidly as the surrounding environment complicates. In this thesis, we thoroughly study a modified version of RRT called the probabilistic RRT (pRRT) for motion planning of autonomous vehicles. The pRRT algorithm reduces the randomness of the standard RRT algorithm and takes into account the destination location and the positions of the obstacles to find a path around the obstacles and toward the destination point. The algorithm was experimentally validated and confirmed the simplistic transfer from simulations to reality. In the second part of the thesis, we propose a novel algorithm that combines the properties of pRRT and another well-known algorithm called RRT-Connect. This algorithm plans collision-free paths from the start, and the goal points towards free space around the obstacles simultaneously and then combines these fragmented paths. This reduces the overall planning time and was found to be better at providing smooth paths.

One of today's most serious social, economical and environmental problems is traffic congestion. In addition to the financial cost of the problem, the number of traffic related injuries and casualties is very high. A recently considered approach to increase safety while reducing congestion and improving driving conditions is Automated Highway Systems (AHS). The AHS will evolve from the present highway system to an intelligent vehicle/highway system that will incorporate communication, vehicle control and traffic management techniques to provide safe, fast and more efficient surface transportation. A key factor in AHS deployment is intelligent vehicle control. While the technology to safely maneuver the vehicles exists, the problem of making intelligent decisions to improve a single vehicle's travel time and safety while optimizing the overall traffic flow is still a stumbling block. We propose an artificial intelligence technique called stochastic learning automata to design an intelligent vehicle path controller. Using the information obtained by on-board sensors and local communication modules, two automata are capable of learning the best possible (lateral and longitudinal) actions to avoid collisions. This learning method is capable of adapting to the automata environment resulting from unmodeled physical environment. Simulations for simultaneous lateral and longitudinal control of an autonomous vehicle provide encouraging results. Although the learning approach taken is capable of providing a safe decision, optimization of the overall traffic flow is also possible by studying the interaction of the vehicles. The design of the adaptive vehicle path planner based on local information is then carried onto the interaction of multiple intelligent vehicles. By analyzing the situations consisting of conflicting desired vehicle paths, we extend our design by additional decision structures. The analysis of the situations and the design of the additional structures are made possible by the study of the interacting reward-penalty mechanisms in individual vehicles. The definition of the physical environment of a vehicle as a series of discrete state transitions associated with a "stationary automata environment" is the key to this analysis and to the design of the intelligent vehicle path controller. This work was supported in part by the Center for Transportation Research and Virginia DOT under Smart Road project, by General Motors ITS Fellowship program, and by Naval Research Laboratory under grant no. N000114-93-1-G022. / Ph. D.

Pursuit-evasion games are the subject of a variety of research initiatives seeking to provide some level of autonomy to mobile, robotic vehicles with on-board controllers. Applications of these controllers include defense topics such as unmanned aerial vehicle (UAV) and unmanned underwater vehicle (UUV) navigation for threat surveillance, assessment, or engagement. Controllers implementing pursuit-evasion algorithms are also used for improving everyday tasks such as driving in traffic when used for collision avoidance maneuvers. Currently, pursuit-evasion tactics are incorporated into the control by solving the Hamilton-Jacobi-Isaacs (HJI) equation explicitly, simplifying the solution using approximate dynamic programming, or using a purely finite-horizon approach. Unfortunately, these methods are either subject to difficulties of long computational times or having no guarantees of succeeding in the pursuit-evasion game. This leads to more difficulties of implementing these tactics on-line in a real robotic scenario where the opposing agent may not be known before the maneuver is required. This dissertation presents a novel method of solving the HJI equation by partitioning the state space into regions of local, finite horizon control laws. As a result, the HJI equation can be reduced to solving the Hamilton-Jacobi-Bellman equation recursively as information is received about an opposing agent. Adding complexity to the problem structure results in a decreased calculation time to allow pursuit-evasion tactics to be calculated on-board an agent during a scenario. The algorithms and implementation methods are given explicitly and illustrated with an example of two robotic vehicles in a collision avoidance maneuver. / Ph. D.

In the field of modern automotive engineering, many researchers are focusing on the development of advanced vehicle control systems such as autonomous vehicle systems and Advanced Driver Assistance Systems (ADAS). Furthermore, Driver Assistance Systems (DAS) such as cruise control, Anti-Lock Braking Systems (ABS), and Electronic Stability Control (ESC) have become widely popular in the automotive industry. Therefore, vehicle control research attracts attention from both academia and industry, and has been an active area of vehicle research for over 30 years, resulting in impressive DAS contributions. Although current vehicle control systems have improved vehicle safety and performance, there is room for improvement for dealing with various situations. The objective of the research is to develop a predictive vehicle control system for improving vehicle safety and performance for autonomous vehicles and ADAS. In order to improve the vehicle control system, the proposed system utilizes information about the upcoming local driving environment such as terrain roughness, elevation grade, bank angle, curvature, and friction. The local driving environment is measured in advance with a terrain measurement system to provide terrain data. Furthermore, in order to obtain the information about road conditions that cannot be measured in advance, this work begins by analyzing the response measurements of a preceding vehicle. The response measurements of a preceding vehicle are acquired through Vehicle-to-Vehicle (V2V) or Vehicle-to-Infrastructure (V2I) communication. The identification method analyzes the response measurements of a preceding vehicle to estimate road data. The estimated road data or the pre-measured road data is used as the upcoming driving environment information for the developed vehicle control system. The metric that objectively quantifies vehicle performance, the Performance Margin, is developed to accomplish the control objectives in an efficient manner. The metric is used as a control reference input and continuously estimated to predict current and future vehicle performance. Next, the predictive control algorithm is developed based on the upcoming driving environment and the performance metric. The developed system predicts future vehicle dynamics states using the upcoming driving environment and the Performance Margin. If the algorithm detects the risks of future vehicle dynamics, the control system intervenes between the driver's input commands based on estimated future vehicle states. The developed control system maintains vehicle handling capabilities based on the results of the prediction by regulating the metric into an acceptable range. By these processes, the developed control system ensures that the vehicle maintains stability consistently, and improves vehicle performance for the near future even if there are undesirable and unexpected driving circumstances. To implement and evaluate the integrated systems of this work, the real-time driving simulator, which uses precise real-world driving environment data, has been developed for advanced high computational vehicle control systems. The developed vehicle control system is implemented in the driving simulator, and the results show that the proposed system is a clear improvement on autonomous vehicle systems and ADAS. / Ph. D. / In the field of modern automotive engineering, many researchers are focusing on the development of advanced vehicle control systems such as autonomous vehicle systems and Advanced Driver Assistance Systems (ADAS). Furthermore, cruise control, Anti-Lock Braking Systems, and Electronic Stability Controls have become widely popular in the automotive industry. Although vehicle control systems have improved vehicle safety and performance, there is still room for improvement for dealing with various situations. The objective of the research is to develop a predictive vehicle control system for improving vehicle safety and performance for autonomous vehicles and ADAS. In order to improve the vehicle control system, the proposed system utilizes information about the upcoming driving conditions such as road roughness, elevation grade, bank angle, and curvature. The driving environment is measured in advance with a terrain measurement system. Furthermore, in order to obtain the information about road conditions that cannot be measured in advance, this work begins by analyzing a preceding vehicle’s response to the road. The combined road data is used as the upcoming driving environment information. The measurement that indicates vehicle performance, the Performance Margin, is developed to accomplish the research objectives. It is used in the developed control system, which predicts future vehicle performance. If the system detects future risks, the control system will intervene to correct the driver’s input commands. By these processes, the developed system ensures that the vehicle maintains stability, and improves vehicle performance regardless of the upcoming and unexpected driving conditions. To implement and evaluate the proposed systems, a driving simulator has been developed. The results show that the proposed system is a clear improvement on autonomous vehicle systems and ADAS.

International Telemetering Conference Proceedings / October 28-31, 1996 / Town and Country Hotel and Convention Center, San Diego, California / The objective of the FIRST project is to define a modern DoD Standard Datalink capability. This defined capability or standard is to provide a solution to wide variety of test and training range digital data radio communications problems with a common set of components, flexible to fit a broad range of applications, yet be affordable in all of them. This capability is to be specially designed to meet the expanding range distances and data transmissions rates needed to test modern weapon systems. Presently, the primary focus of the project is more on software, protocols, design techniques and standards, than on hardware development. Existing capabilities, on going developments and emerging technologies are being investigated and will be utilized as appropriate. Modern processingintensive communications technology can perform many complex range data communications tasks effectively, but a large-scale development effort is usually necessary to exploit it to its full potential. Yet, range communications problems are generally of limited scope, so different from one another that a communication system applicable to all of them is not likely to solve any of them well. FIRST will resolve that dilemma by capitalizing on another feature of modern communications technology: its high degree of programmability. This can enable custom-tailoring of datalink operation to particular applications, just as a PC can be tailored to perform a multitude of diverse tasks, through appropriate selection of software and hardware components.

Hydrogen fuel cell technology has emerged as an efficient and clean alternative to internal combustion engines for powering vehicles, and hydrogen powertrains will aid in addressing key environmental issues such as urban air quality and global warming. This work demonstrates the effectiveness of a „hardware-in-loop‟ (HIL) simulation system for validating the safety and effectiveness of control algorithms for a hydrogen fuel cell hybrid passenger vehicle. A significant amount of the work completed in conjunction with the thesis topic was the design and construction of the fuel cell vehicle for competition. Producing a „rolling test bench‟ that generates data to be used to create HIL simulation models required nearly two years of work before an acceptable level of reliability was reached to produce usable data. Some detail will be given in this thesis regarding the infrastructure modifications required to safely build a hydrogen fuel cell vehicle, as well as the design challenges faced in the integration of a fuel cell power module, two electric drive motors, a nickel metal hydride battery, and required power electronics into a small sport utility vehicle originally designed for an internal combustion powertrain. The virtual control validation performed involved designing dynamic models of the systems of interest and performing real-time simulation to ensure that the appropriate controller response is observed. For this thesis, emphasis was placed on several key vehicle control topics. Communication robustness was evaluated to ensure that the complicated vehicle communication network could effectively handle traffic from the six powertrain sub-controllers. Safety algorithms were tested for appropriate response to fault conditions. Control systems were developed and tuned offline reducing the amount of time required for in-vehicle development and testing. Software-in-the-loop simulation was used to check initial code integrity and to validate the hardware-in-the-loop vehicle models. The methodology presented in this work was found to be sufficient for a thorough safety and rationality evaluation of control strategies for hybrid fuel cell vehicles.