Joshua E. Siegel

Cloudthink: a scalable secure platform for mirroring transportation systems in the cloud

AbstractWe present a novel approach to developing a vehicle communication platform consisting of a low-cost, open-source hardware for moving vehicle data to a secure server, a Web Application Programming Interface (API) for the provision of third-party services, and an intuitive user dashboard for access control and service distribution. The CloudThink infrastructure promotes the commoditization of vehicle telematics data by facilitating easier, flexible, and more secure access. It enables drivers to confidently share their vehicle information across multiple applications to improve the transportation experience for all stakeholders, as well as to potentially monetize their data. The foundations for an application ecosystem have been developed which, taken together with the fair value for driving data and low barriers to entry, will drive adoption of CloudThink as the standard method for projecting physical vehicles into the cloud. The application space initially consists of a few fundamental and importan...

Smartphone-Based Wheel Imbalance Detection

Onboard sensors in smartphones present new opportunities for vehicular sensing. In this paper, we explore a novel application of fault detection in wheels, tires and related suspension components in vehicles. We present a technique for in-situ wheel imbalance detection using accelerometer data obtained from a smartphone mounted on the dashboard of a vehicle having balanced and imbalanced wheel conditions. The lack of observable distinguishing features in a Fourier Transform (FT) of the accelerometer data necessitates the use of supervised machine learning techniques for imbalance detection. We demonstrate that a classification tree model built using Fourier feature data achieves 79% classification accuracy on test data. We further demonstrate that a Principal Component Analysis (PCA) transformation of the Fourier features helps uncover a unique observable excitation frequency for imbalance detection. We show that a classification tree model trained on randomized PCA features achieves greater than 90% accuracy on test data. Results demonstrate that the presence or absence of wheel imbalance can be accurately detected on at least two vehicles of different make and model. Sensitivity of the technique to different road and traffic conditions is examined. Future research directions are also discussed.Copyright

Vehicular engine oil service life characterization using On-Board Diagnostic (OBD) sensor data

Standardized vehicular On-Board Diagnostics (OBD) systems offer access to information commonly used for fault notification and reactive diagnostic services. Recently, there have been efforts to use OBD data to diagnose and predict faults prior to catastrophic failure events. Engine oil service life, a parameter directly related to engine longevity, is difficult to measure conventionally. We show that the rate of engine coolant temperature rise, readily obtainable through the OBD suite, can serve as a proxy to indicate the remaining engine oil life. We demonstrate consistent results for one vehicle under similar environmental and unloaded engine operating conditions. We also examine the validity of this approach under varying environmental and engine loading conditions with tests on a second vehicle.

A Survey of the Connected Vehicle Landscape—Architectures, Enabling Technologies, Applications, and Development Areas

This paper summarizes the state of the art in connected vehicles—from the need for vehicle data and applications thereof, to enabling technologies, challenges, and identified opportunities. Connectivity is increasing around the world and its expansion to vehicles is no exception. With improvements in connectivity, sensing, and computation, the future will see vehicles used as development platforms capable of generating rich data, acting based on inference, and effecting great change in transportation, the human-vehicle dynamic, the environment, and the economy. Connected vehicle technologies have already been used to improve fleet safety and efficiency, with emerging technologies additionally allowing data to be used to inform aspects of vehicle design, ownership, and use. While the demand for connected vehicles and its enabling technology has progressed significantly in recent years, there remain challenges to connected and collaborative vehicle application deployment before the full potential of connected cars may be realized. From extensibility and scalability to privacy and security, this paper informs the reader about key enabling technologies, opportunities, and challenges in the connected vehicle landscape.

Air filter particulate loading detection using smartphone audio and optimized ensemble classification

Abstract Automotive engine intake filters ensure clean air delivery to the engine, though over time these filters load with contaminants hindering free airflow. Today’s open-loop approach to air filter maintenance has drivers replace elements at predetermined service intervals, causing costly and potentially harmful over- and under-replacement. The result is that many vehicles consistently operate with reduced power, increased fuel consumption, or excessive particulate-related wear which may harm the catalyst or damage machined engine surfaces. We present a method of detecting filter contaminant loading from audio data collected by a smartphone and a stand microphone. Our machine learning approach to filter supervision uses Mel-Cepstrum, Fourier and Wavelet features as input into a classification model and applies feature ranking to select the best-differentiating features. We demonstrate the robustness of our technique by showing its efficacy for two vehicle types and different microphones, finding a best result of 79.7% accuracy when classifying a filter into three loading states. Refinements to this technique will help drivers supervise their filters and aid in optimally timing their replacement. This will result in an improvement in vehicle performance, efficiency, and reliability, while reducing the cost of maintenance to vehicle owners.

Smartphone-Based Vehicular Tire Pressure and Condition Monitoring

Proper tire maintenance is key to the safe and efficient operation of vehicles. Without frequent inspection, pressure can drop and tread can wear down, increasing drag and reducing traction. While modern cars make use of Tire Pressure Monitoring Systems (TPMS), these systems may only identify differential changes in pressure and older vehicles lack sensing entirely. To address this problem, we applied mobile phone accelerometer and GPS data to calculate predicted and true wheel rotational frequencies, and therefore infer tire circumference and related pressure or tread depth. Through tree-based classification, we were able to identify from among five tire states with 20% increases or decreases in tire pressure with 80% accuracy using only a mobile phone. The models demonstrated are robust to road surface and the main differentiating element, the GPS/accelerometer velocity ratio, has proven generalizable across vehicle makes, models, and tire sizes.

Engine Misfire Detection with Pervasive Mobile Audio

We address the problem of detecting whether an engine is misfiring by using machine learning techniques on transformed audio data collected from a smartphone. We recorded audio samples in an uncontrolled environment and extracted Fourier, Wavelet and Mel-frequency Cepstrum features from normal and abnormal engines. We then implemented Fisher Score and Relief Score based variable ranking to obtain an informative reduced feature set for training and testing classification algorithms. Using this feature set, we were able to obtain a model accuracy of over 99 % using a linear SVM applied to outsample data. This application of machine learning to vehicle subsystem monitoring simplifies traditional engine diagnostics, aiding vehicle owners in the maintenance process and opening up new avenues for pervasive mobile sensing and automotive diagnostics.

The Future Internet of Things: Secure, Efficient, and Model-Based

The Internet of Things’ (IoT’s) rapid growth is constrained by resource use and fears about privacy and security. A solution jointly addressing security, efficiency, privacy, and scalability is needed to support continued expansion. We propose a solution modeled on human use of context and cognition, leveraging cloud resources to facilitate IoT on constrained devices. We present an architecture applying process knowledge to provide security through abstraction and privacy through remote data fusion. We outline five architectural elements and consider the key concepts of the “data proxy” and the “cognitive layer.” The data proxy uses system models to digitally mirror objects with minimal input data, while the cognitive layer applies these models to monitor the system’s evolution and to simulate the impact of commands prior to execution. The data proxy allows a system’s sensors to be sampled to meet a specified quality of data target with minimal resource use. The efficiency improvement of this architecture is shown with an example vehicle tracking application. Finally, we consider future opportunities for this architecture to reduce technical, economic, and sentiment barriers to the adoption of the IoT.

Real-time Deep Neural Networks for internet-enabled arc-fault detection

Abstract We examine methods for detecting and disrupting electronic arc faults, proposing an approach leveraging Internet of Things connectivity, artificial intelligence, and adaptive learning. We develop Deep Neural Networks (DNNs) taking Fourier coefficients, Mel-Frequency Cepstrum data, and Wavelet features as input for differentiating normal from malignant current measurements. We further discuss how hardware-accelerated signal capture facilitates real-time classification, enabling our classifier to reach 99.95% accuracy for binary classification and 95.61% for multi-device classification, with trigger-to-trip latency under 220 ms . Finally, we discuss how IoT supports aggregate and user-specific risk models and suggest how future versions of this system might effectively supervise multiple circuits.

Automotive Diagnostics as a Service: An Artificially Intelligent Mobile Application for Tire Condition Assessment

Vehicle tires must be maintained to assure performance, efficiency, and safety. Though vehicle owners may monitor tread depth and air pressure, most are unaware of the safety risks of degrading rubber. This paper identifies the need for tire material condition monitoring and develops a densely connected convolutional neural network to identify cracking from smartphone photographs. This model attains an accuracy of 81.2% on cropped outsample images, besting inexperienced humans’ 55% performance. We develop a web service using this model as the basis of an AI-backed “Diagnostics-as-a-Service” platform for online vehicle condition assessment. By encoding knowledge of visual risk indicators into a neural network model operable from a user’s trusted smartphone, we raise awareness of the risk of degraded rubber and improve vehicle safety without requiring specialized operator training.