Social distancing has become a pressing and challenging issue during the Covid-19 pandemic. In a smart cities context, it becomes possible to measure inter-personal distance using networked cameras and computer vision analysis. We deploy a computer vision pipeline based on Retinanet that identifies pedestrians in streaming video frames, then converts their positions to GPS coordinates for distance calculation and further analysis. This processing is applied to nine camera streams at three locations from around Vanderbilt University. We collect 70 hours of baseline distancing data over the course of two weeks, after which time we deploy small behavioral interventions at the three locations aimed at increasing distancing compliance. Another 70 hours of data with the interventions in place will be analyzed against the baseline data to determine if they had an effect on distancing compliance.
New data sources such as video promise to provide insights into how humans navigate urban infrastructure and enable analysis of human-in-the-loop interactions. This work considers the deployment of portable video data collection units to understand human-driver interactions at unsignalized intersections. Specifically, we present preliminary data collection and results that highlight the value of video data in capturing the nuanced interactions of pedestrians with vehicles when navigating urban streets.
The accurate and timely information about parking occupancy and availability has played a crucial role to solve the smart city challenge related to mobility, by helping drivers to save their time and by avoiding waiting to find a space, to move smoothly, or be in traffic. In recent times, there has been growing interest in the use of Big Data and crowd-sourcing data for both research and commercial applications. However, several challenges remain to extract salient information for designing an accurate and timely parking recommendation system (PRS). Differently from the current state of the artwork our PRS extend the application of standard Machine Learning approaches by proposing the application of an additive regression model (Prophet model) fed by parking meters data (parking meters occurrences). The proposed PRS provides timely forecasting until the next month parking occupancy for each different area using different data sources and an additive-based model (Prophet model). The preliminary results related to the forecasting accuracy on a specific area confirmed how the proposed PRS framework is effective and accurate to provide the forecast of parking meters occurrences until the next month, with an R2 score up to 0.51. The obtained results suggest that the proposed approach is a viable solution for providing reliable forecasting of parking occupancy for different areas and different data sources by modeling non-linear, non-periodic, and weekly periodic changes of the parking meter data.
The Controller Area Network (CAN) bus protocol is used in modern vehicles for sharing messages between several control units within a vehicle. CAN bus messages are encoded with unknown scheme and decoding these messages provide unlimited access to valuable information that is used in many autonomous vehicles applications. This paper proposes a ROS based package (CAN-to-ROS) for monitoring, recording, and real-time and offline decoding of CAN bus messages. The package is developed in the ROS framework to add modularity and ease of integration with other software, and it is written in C++ to guarantee speed of the execution during run-time. For realtime decoding of CAN bus data, CAN-to-ROS package used in conjunction with other library called Libpanda that provide access to CAN bus message from a vehicle. The package was evaluated and tested on a Raspberry Pi with real CAN bus data from a Toyota RAV4. The results confirm the capabilities of CAN-to-ROS package and resulted in using the package in other research projects.
Modern intelligent urban mobility applications are underpinned by large-scale, multivariate, spatiotemporal data streams. Working with this data presents unique challenges of data management, processing and presentation that is often overlooked by researchers. Therefore, in this work we present an integrated data management and processing framework for intelligent urban mobility systems currently in use by our partner transit agencies. We discuss the available data sources and outline our cloud-centric data management and stream processing architecture built upon open-source publish-subscribe and NoSQL data stores. We then describe our data-integrity monitoring methods. We then present a set of visualization dashboards designed for our transit agency partners. Lastly, we discuss how these tools are currently being used for AI-driven urban mobility applications that use these tools.
This paper describes an approach to identify undecoded Controller Area Network (CAN) data from one vehicle, based on the data similarity to previously decoded CAN data from another vehicle. Modern vehicles communicate data and signals from on-board sensors and controllers through the CAN bus. Networked sensors contain information such as wheel speeds, fuel gauges, turn signals, and radar signals. In the effort to use this information and make cars safer through human-in-the-loop CPS, signals on the CAN bus such as wheel speed and radar can be used to support the driver. However, data from the CAN bus are encoded and in some cases compressed, and different car manufacturers use different encoding schemes to represent data on the CAN bus. With hundreds of messages and thousands of possible encoding schemes to consider, it is laborious to identify the unique bits and encoding schemes that represent signals on each vehicle. In this study, we propose a method for training a Long Short-Term Memory (LSTM) neural network on known radar signals from one vehicle manufacturer, a Toyota, and successfully apply the network to identify the encoding for radar signals on a different vehicle, a Honda. By augmenting the training dataset with varied encoding bit boundaries, a small and lightweight LSTM network can learn to recognize radar data across different encoding schemes. The results are an improvement on exhaustive-search algorithms and other methods previously used in the search for such signals.
Cyber-Physical Systems (CPS) generally involve time-critical components due to physical dynamics, therefore necessitating high-performance subsystems. This is also true in data collection scenarios to infer physical phenomena. This paper covers Libpanda as an example of a component that has been designed to address performance issues in CPS implementations. Libpanda is a C++ library that interfaces software with a Comma.ai Panda device. Pandas are used for installation in modern vehicles to read the vehicle CAN bus, providing rich sensor data and limited vehicle control through message injection. The motivation to design lib-panda stems from the lack of performance in Python-based code that runs on inexpensive hardware like a Raspberry Pi. In such situations, Python code would result in utilizing 92% CPU while also dropping around 40% of the CAN packet due to bottlenecks. Without using different tools, inconsistent data collection means a loss of time-based vehicle state interpretation. Libpanda addresses these issues through implementation in a different language and implementation of different design paradigms involving asynchronous calls and multithreading. The Panda also features a GPS module that allows multiple instances to synchronize clocks for large-scale data collection scenarios. Libpanda has been designed with time-synchronization in mind to aid in the measurement of inter-vehicle dynamics. The performance improvements of libpanda have resulted in it becoming an important component in automotive dynamics research that requires a higher technical performance in large-scale experiments.
This work presents an integrated framework of: vehicle dynamics models, with a particular attention to instabilities and traffic waves; vehicle energy models, with particular attention to accurate energy values for strongly unsteady driving profiles; and sparse Lagrangian controls via automated vehicles, with a focus on controls that can be executed via existing technology such as adaptive cruise control systems. This framework serves as a key building block in developing control strategies for human-in-the-loop traffic flow smoothing on real highways. In this contribution, we outline the fundamental merits of integrating vehicle dynamics and energy modeling into a single framework, and we demonstrate the energy impact of sparse flow smoothing controllers via simulation results.