Data

Designing traffic-smoothing cruise controllers that can be deployed onto autonomous vehicles is a key step towards improving traffic flow, reducing congestion, and enhancing fuel efficiency in mixed autonomy traffic. We bypass the common issue of having to carefully fine-tune a large traffic micro-simulator by leveraging real-world trajectory data from the I–24 highway in Tennessee, replayed in a one-lane simulation. Using standard deep reinforcement learning methods, we train energy-reducing wave-smoothing policies. As an input to the agent, we observe the speed and distance of only the...

Accurate estimation of vehicle fuel consumption typically requires detailed modeling of complex internal powertrain dynamics, often resulting in computationally intensive simulations. However, many transportation applications-such as traffic flow modeling, optimization, and control-require simplified models that are fast, interpretable, and easy to implement, while still maintaining fidelity to physical energy behavior. This work builds upon a recently developed model reduction pipeline that derives physics-like energy models from high-fidelity Autonomie vehicle simulations. These reduced...

Middleware has emerged as an important architectural component in supporting distributed applications. The role of middleware is to present a unified programming model to application writers and to mask out problems of heterogeneity and distribution. Mobile sensors fall into the space of distributed systems that suffer from isolated data sources, heterogeneous communication infrastructure and varying application requirements. In this report, we provide a middleware architecture that addresses the needs of a distributed system made of mobile sensors in general and discuss the implementation...

Caltrans collects traffic data for many monitoring and control applications and the ultimate goal of the traffic surveillance system is to provide accurate data to these high level applications. The surveillance system includes data measurement, averaging and verification algorithms. This report presents improvements to many elements of the surveillance system. First, section 2addresses many shortcomings in average speed estimation at single loop detectors, as well as other sensors that estimate speed from average flow and occupancy. At the root of these problems is the fact that the...