Advanced Transportation Systems

Abstract

Autonomous transportation technologies must become robust along multiple dimensions before widespread public adoption. The following research challenges of cyber-physical systems must be overcome before autonomous vehicles become practical:

1. (a) addressing the cyber-physical challenges inside the automobile to ensure the timely, safe and reliable operations of sensors, actuators, processors, communication networks, when any could fail at any time, with graceful degradation modes,

2. (b) dealing with wide-ranging physical conditions and uncertainties external to the automobile including pedestrians, bicyclists, motor-bikes, night/dusk/dawn lighting, snow/rainy/dusty conditions, smooth/icy/unpaved roads, highway speeds, road closures, detours and accidents,

3. (c) enabling real-time communications to/from the automobile using vehicle-to-vehicle (V2V) and vehicle- to-infrastructure (V2I) networks to coordinate behaviors with other vehicles and to obtain/react to information about traffic/accident conditions, traffic light and signs, and

4. (d) creating verification and validation technologies, modeling techniques and tools to validate the correctness and robustness of integrated automotive cyber-physical system behaviors.

We seek to make important research contributions in the domains of safety-critical real-time fault- tolerant distributed cyber-physical platforms, end-to-end resource management, cooperative vehicular networks, cyber-physical system modeling and analysis tools, dynamic object detection/recognition, real-time perception and planning algorithms under uncertainty and adverse weather conditions, hybrid systems verification, and safe dynamic behaviors under constantly changing operating conditions. Multiple intermediate capabilities in the form of active safety features will also be enabled.

Societal and Economic Benefits

Automotive accidents result in about 32,000 fatalities and 3 million injuries every year in the United States. About 1.2 million automobile fatalities occur across the globe annually. The global annual cost of road injuries in medical care, disability and property damage is $518 billion. In fact, road traffic injuries represent the leading cause of death worldwide for people ages 10 to 24. Many accidents are due to humans being distracted, unfocused, upset, angry, sad, moody, physically or visually impaired, tired, drowsy or even drunk. They can also underestimate the severity of adverse weather and road conditions. Fully autonomous or even near-autonomous vehicles controlled by ever-vigilant cyber- physical systems can lead over time to significant declines in accidents, and resulting deaths/injuries. In terms of time wasted, the average US driver spends a week stuck in traffic per year. In the European Union alone, 80 billion EU are estimated to be wasted per year due to traffic congestion. Autonomous vehicles over time can offload some driving chores from humans, and make time spent in congestion more productive. Vehicular networks can also help find the best-possible routes to a destination in real- time. As the US population ages, the reaction rates of its senior citizens will slow down, and having actively controlled vehicles will enhance their independence when they cannot drive anymore. Physically or visually challenged citizens will also benefit from these options.

Award ID: 1035813