Brendan Morris

Traffic Prediction Prediction and Visualization

A major concern for traffic operators is the early detection of congestions and incidents. With early detection, resources can be allocated and the incident can be addressed more quickly resulting in significant delay savings, which can improve safety by providing critical moments for emergency personnel. While many tools are currently available, such as traffic maps by companies like Google or reports from Waze, their data is designed for public consumption rather than for traffic operations. The goal of this project is to further the development of a traffic prediction platform which combines traffic sensors with machine learning algorithms for prediction up to one hour into the future with a web-based visualization dashboard for traffic operator use. This project combines data science through use of databases and data pre-processing, machine learning with deep learning-based traffic prediction models, and web-based visualization frameworks.  

Required skills: programming (preferably Python); preferred skills: machine learning

Brendan Morris

Smart Intersection Sensor for Unsafe/Illegal Pedestrian Crossing

Pedestrian safety continues to be a major concern for Las Vegas. Over the past few years an alarming number of pedestrian fatalities have occurred. Many accidents have occurred in places such as Boulder Highway where pedestrians risk crossing many lanes of high-speed traffic with low-illumination to avoid walking the extra mile to a marked crosswalk. The State of Nevada has been examining technologies to help prevent such issues such as LiDAR sensors. While LiDAR is effective, the sensors are expensive, which limits wide-spread deployment. In contrast, camera-based systems are inexpensive and, with modern deep-learning techniques, can accurately detect pedestrians. REU students will develop a complete low-light camera system for detection of unsafe and illegal pedestrian crossings. The students will work with convolutional neural networks (CNNs) to detect and track pedestrians and count the number of illegal crossings. Deep trajectory forecasting techniques will be used to provide a prediction of the intent to cross illegally for advanced warning.

Required skills: programming (preferably Python); preferred skills: computer vision/image processing, machine learning

Sarah Harris

Traffic Signal Information Display

One tool for smoother, more energy efficient driving and improved traffic flow on arterials is the use of signal phase and timing (SPaT) information as generated from a traffic controller. While many intersections in Las Vegas are transmitting SPaT messages, few vehicles use the information as it is relegated to high-end (Audi Traffic Light Information) or self-driving vehicles. The purpose of this project is to develop a low-cost SPaT receiver and display to indicate the amount of time left for the signal to turn green, yellow, or red and to provide speed guidance to minimize wait time at an intersection. The REU students will explore V2X communication protocols to design and develop a radio receiver that can be paired with a Raspberry PI as a lightweight display device. The display will be tested at Las Vegas intersections to characterize the performance in terms of phase accuracy, delay, and distance.

Beiyu Lin

Hearing Loss Detection in Smart Environment

Sound is an important modality to perceive and understand environments. With the development of digital technology, massive amounts of smart devices in use everywhere (e.g., smartphones and autonomous robots) can collect sound data. Approximately 15% of Americans aged 18 and above have some trouble hearing. Over six million (2.5%) Americans aged 12 or over have severe-to-profound hearing loss in at least one ear. This project focuses on hearing loss detection in a smart environment in order to provide early detection and intervention.

Grzegorz Chmaj

Smart Parking Optimization

Parking lots and garages, especially in the areas where large numbers of vehicles arrive at the same time, have problems with traffic congestion. The goal of this project is to develop an IoT system to provide information about free parking spots and also to optimize the traffic in parking lots and garages, targeting one of the crucial needs of a Smart City. The key components of the system are the driver’s smartphone, the parking app and cloud service to collect and interchange the data (IoT architecture). The mobile app will monitor the driver’s behavior and detect when the driver is approaching the car or driving towards the spot. This information will be sent to the cloud, analyzed result will be shared with drivers who are searching for a parking spot. The system will be able to provide the information about current occupancy in the parking lot, predict parking spots to be freed soon and provide live information about the closest free parking spots. Areas utilized in this project are: programming, mobile programming, cloud technologies, databases and machine learning.
Goals of the project are:
#1 predict free parking spots based on drivers’ movements.
#2 include multiple parking lots.

Required skills: programming (preferably Python); preferred skills: machine learning

Grzegorz Chmaj

Shared Mobility in Smart Cities

Smart cities give many opportunities to optimize traffic and decrease congestion. Recent growing popularity of shared means of transport open new possibilities for the people to get to their destination fast. This project considers a smart city with multiple shared mobility options: 1) shared cars 2) shared bicycles 3) shared scooters. Each of these vehicles has different speed and can be rented from the pickup/dropoff location using smartphone app – commuter knows their current availability / locations. Taking advantage of multiple sharing transport options, commuter’s travel time between two locations needs to be minimized. Shared cars are the fastest by speed, but when they face heavy traffic then the travel time significantly increases – so the proposed system should be able to predict heavy traffic and plan the switch (or multiple switches) to another type of shared transportation.
Goal of the system is to build and test the ML model optimizing the travel time between two locations, using multiple types of shared mobility vehicles.

Required skills: programming (preferably Python); preferred skills: machine learning.

Venkatesan Muthukumar

Detection of Acoustic Events in Urban Environment

Detection of acoustic events in urban environments such as vehicle accidents, crowd behaviors, gun shots, etc. are of significant importance in developing safer and smarter cities. Traditionally, these activities are detected based on video/image processing methods. However, given the complex urban spatial constraints, such as congested and occluded roads and privacy issues, video/image based methods have not been pursued. An alternate form of such event detection is to use acoustic signals. However, acoustic methods pose their own challenges, such as background noise, and polyphonic signals make event detection difficult. This work will focus on real-time detection of acoustic events in real urban environments. The REU students will interface a digital microphone to an IoT enabled hardware system and use the python programming language to collect and store various urban acoustic data. This data will be stored in the cloud and then, using cloud tools, audio features will be extracted and machine learning deployed for classification and detection of various urban events – specifically unsafe and critical events.

Venkatesan Muthukumar

Enrich Traffic Information with Real Time & Historical Vehicle Data

Autonomous vehicles and semi-autonomous vehicles capture continuous data from various sensors about the surrounding environment. This data is archived by various third-party data warehouse companies. These data are available for self-driving cars in real-time to select optimized routes to mitigate congestion and minimize travel time. In this project, we explore API to collect and visualize the above real-time data, dynamically update traffic patterns, and perform data analytics for optimal routing.

Requirement: Python, JS, etc.