Navigation, Guidance, Control, and Dynamics Commons^™

Planning Algorithms Under Uncertainty For A Team Of A Uav And A Ugv For Underground Exploration, Matteo De Petrillo

Graduate Theses, Dissertations, and Problem Reports (ETD)

Robots’ autonomy has been studied for decades in different environments, but only recently, thanks to the advance in technology and interests, robots for underground exploration gained more attention. Due to the many challenges that any robot must face in such harsh environments, this remains an challenging and complex problem to solve.

As technology became cheaper and more accessible, the use of robots for underground ex- ploration increased. One of the main challenges is concerned with robot localization, which is not easily provided by any Global Navigation Services System (GNSS). Many developments have been achieved for indoor mobile ground robots, making …

Efficient End-To-End Autonomous Driving, Hesham Eraqi

Theses and Dissertations

Steering a car through traffic is a complex task that is difficult to cast into algorithms. Therefore, researchers turn to train artificial neural networks from front-facing camera data stream along with the associated steering angles. Nevertheless, most existing solutions consider only the visual camera frames as input, thus ignoring the temporal relationship between frames. In this work, we propose a Convolution Long Short-Term Memory Recurrent Neural Network (C-LSTM), which is end-to-end trainable, to learn both visual and dynamic temporal dependencies of driving. Additionally, We introduce posing the steering angle regression problem as classification while imposing a spatial relationship between the …

Towards Sensorimotor Coupling Of A Spiking Neural Network And Deep Reinforcement Learning For Robotics Application, Kashu Yamazaki

Mechanical Engineering Undergraduate Honors Theses

Deep reinforcement learning augments the reinforcement learning framework and utilizes the powerful representation of deep neural networks. Recent works have demonstrated the great achievements of deep reinforcement learning in various domains including finance,medicine, healthcare, video games, robotics and computer vision.Deep neural network was started with multi-layer perceptron (1stgeneration) and developed to deep neural networks (2ndgeneration)and it is moving forward to spiking neural networks which are knownas3rdgeneration of neural networks. Spiking neural networks aim to bridge the gap between neuroscience and machine learning, using biologically-realistic models of neurons to carry out computation. In this thesis, we first provide a comprehensive review …

Evaluating Driving Performance Of A Novel Behavior Planning Model On Connected Autonomous Vehicles, Keyur Shah

Honors Scholar Theses

Many current algorithms and approaches in autonomous driving attempt to solve the "trajectory generation" or "trajectory following” problems: given a target behavior (e.g. stay in the current lane at the speed limit or change lane), what trajectory should the vehicle follow, and what inputs should the driving agent apply to the throttle and brake to achieve this trajectory? In this work, we instead focus on the “behavior planning” problem—specifically, should an autonomous vehicle change lane or keep lane given the current state of the system?

In addition, current theory mainly focuses on single-vehicle systems, where vehicles do not communicate with …

Hardware Security Of The Controller Area Network (Can Bus), David Satagaj

Senior Honors Theses

The CAN bus is a multi-master network messaging protocol that is a standard across the vehicular industry to provide intra-vehicular communications. Electronics Control Units within vehicles use this network to exchange critical information to operate the car. With the advent of the internet nearly three decades ago, and an increasingly inter-connected world, it is vital that the security of the CAN bus be addressed and built up to withstand physical and non-physical intrusions with malicious intent. Specifically, this paper looks at the concept of node identifiers and how they allow the strengths of the CAN bus to shine while also …

Dashboard-Mounted Driver Monitoring System, Daniel Lein, Prahlad Sarma, William Roussell, Nathan Nuñez, Elvis Manzanares

ENGS 89/90 Reports

The objective of this project is to deliver safe DMS hardware and interfacing/preprocessing software on which Toyota can run its drowsiness detection neural net. The system shall: securely house all necessary imaging, electronic, and computer hardware to output facial recognition data to the car’s computer, without data input from any other car system; obscure less than 10% of the driver’s field of view out of a standard windshield; be able to capture and process images such that the image classification neural net recognize the images, meeting the top-1 benchmark accuracy statistic across static and dynamic car-cabin lighting conditions; and meet …

Techniques To Solve Decision-Making Problems, Dilnoz Tulkunovna Muhamediyeva, Bekmuratov Fayzievich Tulkun

Chemical Technology, Control and Management

Solving decision-making problems in poorly formalized systems only with the help of deterministic and probabilistic methods is insufficient. To do this, it is necessary to widely apply the methods of hybrid intelligent systems and, especially, the methods of “soft” calculations (SoftCalculation, SoftComputing) and the directions of ComputationalIntelligence — intelligent computing technologies that are emerging on this theoretical and methodological base. An immune - fuzzy algorithm for the synthesis of fuzzy inference systems (FIS) is proposed. A two-stage adaptive FIS synthesis algorithm is described. At the first stage, the initial fuzzy parameters are clustered in order to reduce the number of …

Quantitative Performance Assessment Of Lidar-Based Vehicle Contour Estimation Algorithms For Integrated Vehicle Safety Applications, David M. Mothershed

College of Graduate Studies: Theses & Dissertations

Many nations and organizations are committing to achieving the goal of `Vision Zero' and eliminate road traffic related deaths around the world. Industry continues to develop integrated safety systems to make vehicles safer, smarter and more capable in safety critical scenarios. Passive safety systems are now focusing on pre-crash deployment of restraint systems to better protect vehicle passengers. Current commonly used bounding box methods for shape estimation of crash partners lack the fidelity required for edge case collision detection and advanced crash modeling. This research presents a novel algorithm for robust and accurate contour estimation of opposing vehicles. The presented …

Zips Racing Electric Can Communications, Andrew Jordan, Adam Long, Susanah Kowalewski, Rami Nehme

Williams Honors College, Honors Research Projects

The CAN protocol has been a standard of electronic communication networks of automotive vehicles since the early 2000s due to its robust reliability in harsh environments. For the 2020 competition year, the Zips Racing Electric design team will be building an entirely new, fully-electric vehicle with CAN communication implemented rather than communicating via pure analog signals. Hardware and software can be utilized to read analog electrical signals from a source, such as accelerator and brake sensors, and encode them into a digital message that meets the CAN 2.0B communication protocol standard. Likewise, software can be used to extract data from …

Route Planning For Long-Term Robotics Missions, Christopher Alexander Arend Tatsch

Graduate Theses, Dissertations, and Problem Reports (ETD)

Many future robotic applications such as the operation in large uncertain environment depend on a more autonomous robot. The robotics long term autonomy presents challenges on how to plan and schedule goal locations across multiple days of mission duration. This is an NP-hard problem that is infeasible to solve for an optimal solution due to the large number of vertices to visit. In some cases the robot hardware constraints also adds the requirement to return to a charging station multiple times in a long term mission. The uncertainties in the robot model and environment require the robot planner to account …

Self-Driving Toy Car Using Deep Learning, Fahim Ahmed, Suleyman Turac, Mubtasem Ali

Publications and Research

Our research focuses on building a student affordable platform for scale model self-driving cars. The goal of this project is to explore current developments of Open Source hardware and software to build a low-cost platform consisting of the car chassis/framework, sensors, and software for the autopilot. Our research will allow other students with low budget to enter into the world of Deep Learning, self-driving cars, and autonomous cars racing competitions.

Comparison Of Lqr And Lqr-Mrac For Linear Tractor-Trailer Model, Kevin Richard Gasik

Master's Theses

The United States trucking industry is immense. Employing over three million drivers and traveling to every city in the country. Semi-Trucks travel millions of miles each week and encompass roads that civilians travel on. These vehicles should be safe and allow efficient travel for all. Autonomous vehicles have been discussed in controls for many decades. Now fleets of autonomous vehicles are beginning their integration into society. The ability to create an autonomous system requires domain and system specific knowledge. Approaches to implement a fully autonomous vehicle have been developed using different techniques in control systems such as Kalman Filters, Neural …

Autonomous Watercraft Simulation And Programming, Nicholas J. Savino

Student Scholar Showcase

Automation of various modes of transportation is thought to make travel more safe and efficient. Over the past several decades, advances to semi-autonomous and autonomous vehicles have led to advanced autopilot systems on planes and boats, and an increasing popularity of self-driving cars. We predicted the motion of an autonomous vehicle using simulations in Python. The simulation models the motion of a small scale watercraft, which can then be built and programmed using an Arduino Microcontroller. We examined different control methods for a simulated rescue craft to reach a target. We also examined the effects of different factors, such as …

Edge Heterogeneous Hardware Evaluation Based On Real Connected And Autonomous Vehicles (Cavs) Workloads, Mustafa Ahmad

Research Opportunities for Engineering Undergraduates (ROEU) Program 2018-19

There has recently been a wide expansion of hardware to assist in autonomous driving tasks. On this project, we focus on using some state-of-the-art deep learning workloads in connected autonomous vehicle (CAV) scenarios,such as object detection and object tracking to evaluate the heterogeneous hardware.

Roadnet: Robust Adaptive Network For Information Diffusion In Vanet, Priyashraba Misra

Graduate Theses, Dissertations, and Problem Reports (ETD)

The automotive industry has changed more in the last one decade than ever before. Rapid advancements in autonomous driving have opened up opportunities for CAVs (Connected and Automated Vehicles). Vehicles today rely on a sensor-suite to map the surrounding and use that information for safety and navigation. The sensor's view is limited to its line of sight and this drawback can be tapered off by using Vehicle to Vehicle (V2V) and Vehicle to Infrastructure (V2I), generally referred to as Vehicle to Everything (V2X) communication. In this thesis, we specifically focus on utilizing V2V communication using on-board Dedicated Short Range Communication …

Unsupervised Feature Learning For Point Cloud By Contrasting And Clustering With Graph Convolutional Neural Network, Ling Zhang

Dissertations and Theses

Recently, deep graph neural networks (GNNs) have attracted significant attention for point cloud understanding tasks, including classification, segmentation, and detection. However, the training of such deep networks still requires a large amount of annotated data, which is both expensive and time-consuming. To alleviate the cost of collecting and annotating large-scale point cloud datasets, we propose an unsupervised learning approach to learn features from unlabeled point cloud ”3D object” dataset by using part contrasting and object clustering with GNNs. In the contrast learning step, all the samples in the 3D object dataset are cut into two parts and put into a …

Assessing The Competing Characteristics Of Privacy And Safety Within Vehicular Ad Hoc Networks, Jacob W. Connors

Theses and Dissertations

The introduction of Vehicle-to-Vehicle (V2V) communication has the promise of decreasing vehicle collisions, congestion, and emissions. However, this technology places safety and privacy at odds; an increase of safety applications will likely result in the decrease of consumer privacy. The National Highway Traffic Safety Administration (NHTSA) has proposed the Security Credential Management System (SCMS) as the back end infrastructure for maintaining, distributing, and revoking vehicle certificates attached to every Basic Safety Message (BSM). This Public Key Infrastructure (PKI) scheme is designed around the philosophy of maintaining user privacy through the separation of functions to prevent any one subcomponent from identifying …

Vehicle-To-Barrier Communication During Real-World Vehicle Crash Tests, Samil Temel, Mehmet C. Vuran, Mohammad M.R. Lunar, Zhongyuan Zhao, Abdul Salam, Ronald K. Faller, Cody S. Stolle

School of Computing: Faculty Publications

Vehicle-to-barrier (V2B) communication is expected to facilitate wireless interactions between vehicles and roadside barriers in next-generation intelligent transportation systems. V2B systems will help mitigate single-vehicle, run-off-road crashes, which account for more than 50% of roadside crash fatalities. In this work, the characteristics of the wireless channel prior to and during a crash are analyzed using orthogonal frequency division multiplexing (OFDM) techniques, which has been used in existing vehicular communication systems. More specifically, the performance of OFDM-based V2B links are measured in real-world crash tests for the first time. Three crash tests conducted at the Midwest Roadside Safety Facility, Lincoln, Nebraska, …

Crop Height Estimation With Unmanned Aerial Vehicles, Carrick Detweiler, David Anthony, Sebastian Elbaum

School of Computing: Faculty Publications

An unmanned aerial vehicle (UAV) can be configured for crop height estimation. In some examples, the UAV includes an aerial propulsion system, a laser scanner configured to face downwards while the UAV is in flight, and a control system. The laser scanner is configured to scan through a two-dimensional scan angle and is characterized by a maxi mum range. The control system causes the UAV to fly over an agricultural field and maintain, using the aerial propulsion system and the laser scanner, a distance between the UAV and a top of crops in the agricultural field to within a programmed …

An Empirical Study To Investigate The Effect Of Air Density Changes On The Dsrc Performance, Mostafa El-Said, Vijay Bhuse, Alexander Arendsen

Peer-Reviewed Publications

The primary role of Intelligent Transportation Systems (ITS) system is to implement Advanced Driver Assistance Services (ADAS) such as pedestrian detection, fog detection and collisions avoidance. These services rely on detecting and communicating the environment conditions such as heavy rain or snow with nearby vehicles to improve the driver's visibility. ITS systems rely on DSRC to communicate this information via a Vehicle-to-Vehicle (V2V) or Vehicle-to-Infrastructure (V2I) communications architectures. DSCR performance may be susceptible to environmental changes such as air density, gravitation (gravitational acceleration), air temperature, atmospheric pressure, humidity, and precipitation.

The goal of this research is to investigate whether the …