This paper introduces a novel method for real-time rainfall rate determination in a given region, leveraging signals from connected vehicles as primary data sources. Traditional methods suffer from low temporal resolution and accessibility issues, prompting the need for an alternative approach. Our method utilizes vehicle sensor data, including wiper speed, temperature, humidity, and pressure, alongside ground truth data from Next Generation Weather Radars (NEXRAD). Through a two-stage process, a model is developed to predict the upper bound of rainfall rate based on aggregated vehicle signals. Validation of the model demonstrates its reliability and accuracy, with implications for enhancing advanced driver-assistance systems (ADAS) and ensuring safe driving in adverse weather conditions. This paper outlines the methodology, validation process, and potential applications of the proposed method, offering a practical solution for real-time rainfall prediction using existing connected vehicle infrastructure.

Nowadays, Mobile Crowdsourcing (MCS) systems, which leverage crowdsourced data to enable large-scale sensing, has become a popular research area. A persistent question has been whether the number of deployed sensing assets is sufficient to provide enough spatial and temporal coverage in particular areas and whether there is sufficient data to enable individual crowdsensing use cases within an area. We present a framework to provide such analysis. We also propose a pipeline for processing millions of vehicle telemetry data points into coverage assessment results. A set of novel geospatial algorithms for data processing are presented as well. Our analysis models support multiple types of coverage analysis including Segment mode, Node mode, Unique Mileage mode, and Categorized Road mode. Analysis table results and heatmap results are presented in the Demonstration section.

In this paper, we develop a set of normalized spatial temporal properties from which to assess road performance and detect abnormal traffic stream environments. Data is provided by non-instrumented retail vehicle telemetry data, which captures timestamps, GPS locations, and velocity provided every 3 seconds. Open Street Map (OSM) roadways are subdivided into 500-meter segments. Vehicles traveling across those segments are crowd-sensed every 15 minutes. The demonstration compares two traffic streams over the same segments and time periods exactly two weeks apart and illustrates the effectiveness of the approach. The novelty includes methods to create comparable spatial temporal frames and properties by transformations of telemetry data into driver experiences. The outcome of our approach generates a space time grid of cells from which to detect, describe, and diagnose traffic environments.

Precise positioning of an automobile to within lane-level precision can enable better navigation and context-awareness. However, GPS by itself cannot provide such precision in obstructed urban environments. In this paper, we present a system called CARLOC for lane-level positioning of automobiles. CARLOC uses three key ideas in concert to improve positioning accuracy: it uses digital maps to match the vehicle to known road segments; it uses vehicular sensors to obtain odometry and bearing information; and it uses crowd-sourced location of estimates of roadway landmarks that can be detected by sensors available in modern vehicles. CARLOC unifies these ideas in a probabilistic position estimation framework, widely used in robotics, called the sequential Monte Carlo method. Through extensive experiments on a real vehicle, we show that CARLOC achieves sub-meter positioning accuracy in an obstructed urban setting, an order-of-magnitude improvement over a high-end GPS device.

Precise positioning of an automobile to within lane-level precision can enable better navigation and context-awareness. However, GPS by itself cannot provide such precision in obstructed urban environments. In this paper, we present a system called CARLOC for lane-level positioning of automobiles. CARLOC uses three key ideas in concert to improve positioning accuracy: it uses digital maps to match the vehicle to known road segments; it uses vehicular sensors to obtain odometry and bearing information; and it uses crowd-sourced location of estimates of roadway landmarks that can be detected by sensors available in modern vehicles. CARLOC unifies these ideas in a probabilistic position estimation framework, widely used in robotics, called the sequential Monte Carlo method. Through extensive experiments on a real vehicle, we show that CARLOC achieves sub-meter positioning accuracy in an obstructed urban setting, an order-of-magnitude improvement over a high-end GPS device.

Automotive apps can improve efficiency, safety, comfort, and longevity of vehicular use. These apps achieve their goals by continuously monitoring sensors in a vehicle, and combining them with information from cloud databases in order to detect events that are used to trigger actions (e.g., alerting a driver, turning on fog lights, screening calls). However, modern vehicles have several hundred sensors that describe the low level dynamics of vehicular subsystems, these sensors can be combined in complex ways together with cloud information. Moreover, these sensor processing algorithms may incur significant costs in acquiring sensor and cloud information. In this paper, we propose a programming framework called CARLOG to simplify the task of programming these event detection algorithms. CARLOG uses Datalog to express sensor processing algorithms, but incorporates novel query optimization methods that can be used to minimize bandwidth usage, energy or latency, without sacrificing correctness of query execution. Experimental results on a prototype show that CARLOG can reduce latency by nearly two orders of magnitude relative to an unoptimized Datalog engine.