For a little over the past decade since the DARPA Grand Challenge in 2004 and the more successful Urban Challenge in 2007 autonomous vehicles have seen a surge in popularity with car manufacturers, and companies such as Google and Uber. Light Detection And Ranging (LiDAR) has been one of the major sensors in use to sense for acting on the surrounding environment instead of the classic radar which has a much narrower field of vision. However the cost of the higher end 3D LiDAR systems which started seeing use during the DARPA challenges still have the high cost of $70,000 a piece which is an issue when trying to design a consumer friendly system on a family car. This work aims to investigate alternate 2D LiDAR systems to the costly systems currently in use in many prototypes to find a cost efficient alternative that can detect and track obstacles in front of a vehicle. The introduction begins by summarizing some related prior works, particularly papers from after the Grand Challenge as well as some about the competition itself. Detection and tracking methods for point clouds generated by the LiDAR are explored including ways to search through the data in an efficient manner to meet real-time constraints. Some of the trade-offs in going from a 3D system to a 2D system and examined along with how some of the drawbacks can be mitigated.

In the future autonomous vehicles are expected to navigate independently and manage complex traffic situations. This thesis is one of two theses initiated with the aim of researching which methods could be used within the field of autonomous vehicles. The purpose of this thesis was to investigate how Model Predictive Control could be used in the field of autonomous vehicles. The tasks to generate a safe and economic path, to re-plan to avoid collisions with moving obstacles and to operate the vehicle have been studied. The algorithm created is set up as a hierarchical framework defined by a high and a low level planner. The objective of the high level planner is to generate the global route while the objectives of the low level planner are to operate the vehicle and to re-plan to avoid collisions. Optimal Control problems have been formulated in the high level planner for the use of path planning. Different objectives of the planning have been investigated e.g. the minimization of the traveled length between the start and the end point. Approximations of the static obstacles' forbidden areas have been made with circles. A Quadratic Programming framework has been set up in the low level planner to operate the vehicle to follow the high level pre-computed path and to locally re-plan the route to avoid collisions with moving obstacles. Four different strategies of collision avoidance have been implemented and investigated in a simulation environment.

Autonomous vehicles are an emerging trend in robotics, seen in a vast range of applications and environments. Consequently, Volvo Construction Equipment endeavour to apply the concept of autonomous vehicles onto one of their main products. In the company’s Autonomous Machine project an autonomous wheel loader is being developed. As an ob jective given by the company; a demonstration proving the possibility of conducting a fully autonomous load and haul cycle should be performed. Conducting such cycle requires the vehicle to be able to localise itself in its task space and navigate accordingly. In this Master’s Thesis, methods of solving those requirements are proposed and evaluated on a real wheel loader. The approach taken regarding localisation, is to apply sensor fusion, by extended Kalman filtering, to the available sensors mounted on the vehicle, including; odometric sensors, a Global Positioning System receiver and an Inertial Measurement Unit. Navigational control is provided through an interface developed, allowing high level software to command the vehicle by specifying drive paths. A path following controller is implemented and evaluated. The main objective was successfully accomplished by integrating the developed localisation and navigational system with the existing system prior this thesis. A discussion of how to continue the development concludes the report; the addition of a continuous vision feedback is proposed as the next logical advancement.

Autonomous vehicles have many application possibilities within many different fields like rescue missions, exploring foreign environments or unmanned vehicles etc. For such system to navigate in a safe manner, high requirements of reliability and security must be fulfilled. This master's thesis explores the possibility to use the machine learning algorithm convolutional network on a robotic platform for autonomous path following. The only input to predict the steering signal is a monochromatic image taken by a camera mounted on the robotic car pointing in the steering direction. The convolutional network will learn from demonstrations in a supervised manner. In this thesis three different preprocessing options are evaluated. The evaluation is based on the quadratic error and the number of correctly predicted classes. The results show that the convolutional network has no problem of learning a correct behaviour and scores good result when evaluated on similar data that it has been trained on. The results also show that the preprocessing options are not enough to ensure that the system is environment dependent.

An autonomous vehicle is a vehicle that is capable of navigating and driving with no human intervention whatsoever through the utilization of various sensors and positioning systems. The possible applications of autonomous vehicles are widespread, ranging from the aerospace industry to the mining and military sectors where the exposure of human operators to the operating conditions is hazardous to their health and safety. Automobile accidents have become the leading cause of death in certain segments of the world population. Removing the human driver from the decision-making process through automation may result in significantly safer highways. Although full autonomy may be the ultimate goal, there is huge scope for systems that aid the driver in decision making or systems that take over from the driver under conditions where the human driver fails. The aim of the longitudinal control system to be implemented on the Land Rover test vehicle in this study is to improve the vehicle’s safety by controlling the vehicle’s longitudinal behaviour. A common problem with sports-utility-vehicles is the low rollover threshold, due to a high centre of gravity. Rather than modifying the vehicle to increase the rollover threshold, the aim of the control system presented here is to prevent the vehicle from exceeding speeds that would cause the vehicle to reach its rollover threshold. In order to develop a control system that autonomously controls the longitudinal degree of freedom, a model of the test vehicle (a 1997 Land Rover Defender 110 Wagon) was developed in MSC.ADAMS/View and validated experimentally. The model accurately captures the response of the test vehicle to supply forces as generated by the engine and demand forces applied through drag, braking and engine braking. Furthermore, the model has been validated experimentally to provide reliable simulation results for lateral and vertical dynamics. The control system was developed by generating a reference speed that the vehicle must track. This reference speed was formulated by taking into account the vehicle’s limits due to lateral acceleration, combined lateral and longitudinal acceleration and the vehicle’s performance capabilities. The control system generates the desired throttle pedal position, hydraulic pressure in the brake lines, clutch position and gear selection as output. The MSC.ADAMS\View model of the test vehicle was used to evaluate the performance of the control system on various racetracks of which the GPS coordinates were available. The simulation results indicate that the control system performs as expected. Finally, the control system was implemented on the test vehicle and the performance was evaluated by conducting field tests in the form of a severe double lane change manoeuvre. The results of the field tests indicated that the control system limited the acceleration vector of the vehicle’s centre of gravity to prescribed limits, as predicted by the simulation results. / Dissertation (MEng)--University of Pretoria, 2013. / gm2014 / Mechanical and Aeronautical Engineering / unrestricted

Healthcare visits fall into the category of either Emergency or scheduled visits. Post-pandemic, a significant number of people are avoiding their scheduled and routine healthcare visits due to perceived risk of infection. Any delay in treatments could potentially lead to their condition worsening. The demographic of people avoiding their healthcare include people with reduced mobility ( wheelchair or crutch users) - who experienced a shortage of accessible transportation, people with per-existing conditions who were in-fact the most susceptible to contagious diseases, and lastly unpaid adult care takers. Our current mobility options, which include private vehicles, shared services such as Uber and public transport fail to cater to the unique accessibility and hygiene requirements of patients. This lead me to ask if it was possible to address this gap and create a new solution for a very specific purpose ? My research included understanding the pain points of users with a varying degree of mobility, such as the elderly, who found it challenging to enter vehicle with low ceiling and wheelchair user, who found transferring to the car seat tiring or sometimes impossible without assistance. Users also described how frequently touched surfaces, small enclosed areas with poor ventilation and crowding present in shared/public transportation was a cause of concern and lead to anxiety. During my process, I had used a combination of 3D models, tape drawing and projection mapping to ideate, validate and arrive at the final result. The end product was Neon- a non emergecy on demand mobility service designed to bridge the gap between home and healthcare. It is designed to have a minimal footprint on the road (smaller than a compact car), without compromising comfort of the passengers. Neon has a high ceiling and low floor, so that it’s easy for people with limited mobility to enter and exit the vehicle. In the interior, seats are designed to face each other, in order to provide comfort and security for the patient and also enable the caregiver accompanying them. Neon is wheelchair accessible, with an ejecting ramp that meets at curb height. At the center of the vehicle, a rotating pedestal orients the wheelchair user to their seats, and an innovative locking mechanism secures them in place. During their journey to the healthcare center, patients can have an on-route check up with the medical staff. Depending on the condition of the patient, the medical staff can activate the ‘Emergency mode’, where a blue cross illuminates on the exterior. With its many hygiene and accessibility features, Neon provides a seamless experience for the first mile of healthcare.

This degree project investigates social and technological aspects of human-vehicle interactions with regards to driverless autonomous utilitarian vehicles in urban context and proposes the use of LED lighting as an external Human Machine Interface for vehicle pedestrian signaling.

With autonomy becoming a reality for passenger cars, developing an autonomous for tractor-trailers is the next step for driverless roads. Tractor-trailers are heavy, large, and have a pivot joint between the tractor and trailer that makes the movement between the two more complicated. The purpose of the research presented here is to determine the best forward "looking" perception sensor that will see far out enough for the vehicle to stop in time to avoid hitting an object. In order to determine the best sensor, a review of previous sensors and autonomous vehicle sensors will be explored along with the various perception technology. Additionally, a simulation of a tractor-trailer stopping was created to determine the range necessary for a forward perception sensor and when jackknifing may occur. The best brake type for a tractor-trailer will be recommended as well. Finally, the best forward sensor and senor layout for an autonomous tractor-trailer is made based upon the simulation results for the stopping distance of a tractor-trailer. The work, however, is not fully complete. A discussion of the future work and validation of the sensors selected will give future research goals. / Master of Science / With autonomy becoming a reality for passenger cars, developing an autonomous for tractor-trailers is the next step for driverless roads. Tractor-trailers are heavy, large, and have a pivot joint between the tractor and trailer that makes the movement between the two more complicated. The purpose of the research presented here is to determine the best forward “looking” perception sensor that will see far out enough for the vehicle to stop in time to avoid hitting an object. In order to determine the best sensor, a review of previous sensors and autonomous vehicle sensors will be explored along with the various perception technology. Additionally, a simulation of a tractor-trailer stopping was created to determine the range necessary for a forward perception sensor and when jackknifing may occur. The best brake type for a tractor-trailer will be recommended as well. Finally, the best forward sensor and senor layout for an autonomous tractor trailer is made based upon the simulation results for the stopping distance of a tractor-trailer. The work, however, is not fully complete. A discussion of the future work and validation of the sensors selected will give future research goals.

Collision avoidance is treated as a game of two players with opposing desiderata. In the application to automated car-like vehicles, we will use a differential game in order to model and assess a worst-case analysis. The end result will be an almost analytic representation of a boundary between a "safe" set and a "unsafe" set. We will generalize the research in [27] to non-identical players and begin the setup of the boundary construction. Then we will consider the advantages and disadvantages of manipulation of the cost function through the solution and control techniques. In particular, we introduce a possible way to incorporate a secondary objective such as sticking to a straight path. We also look a hybrid technique to reduce steering when the opposing player is out of the reach of the vehicle; i.e., is out of the "unsafe" set and less extreme maneuvers may be desired. We first look at a terminal cost formulation and through retrograde techniques may shape this boundary between the "safe" and "unsafe" set. We would like this research, or part thereof, to be assessed and simulated on a simulation vehicle such as that used in the Flexible Low-cost Automated Scaled Highway (FLASH) at the Virginia Tech Transportation Institute (VTTI). In preparation, we briefly look at the sensor demands from this game-theoretic approach. / Master of Science

The Flexible Low-cost Automated Scaled Highway (FLASH) laboratory at the Virginia Tech Transportation Institute (VTTI) is one of many facilities dedicated to the field of Intelligent Transportation Systems (ITS). The goal of the FLASH lab is to provide small-scale development and implementation of autonomous control strategies for today's vehicles. The current controller used on the scale vehicles is based solely on the kinematics of the system. This body of work was aimed to develop a dynamic control law to enhance the performance of the existing kinematic controller. This control system is intended to automatically maintain the vehicle's alignment on the road as well as keep the speed of the vehicle constant. Implementation of such systems could conceivably reduce driver fatigue by removing nearly all the burden of the driving process from the driver while on the highway. System dynamics of car-like robots with nonholonomic constraints were employed in this research to create a controller for an autonomous path following vehicle. The application of working kinematic and dynamic models describing car-like robotic systems allowed the development of a nonlinear controller. Simulations of the vehicle and controller were done using MATLAB. Comparisons of the kinematic controller and the dynamic controller presented here were also done. In order to make the simulations model the actual system more closely, measures were taken to approximate actual sensor readings. / Master of Science