Autonomous Vehicles and Urban Mobility
12/08/2009 32"124Emilio Frazzoli, Associate Professor of Aeronautics and AstronauticsDescription: If you had half a million dollars, would you opt for a passenger car that could drive itself (called an autonomous vehicle) or would you choose a new Ferrari? Emilio Frazzoli provides a number of reasons why autonomous vehicles might be the preferred choice, if not the typical one. Autonomous vehicles, that use electronics in place of human drivers, will offer many improvements for urban mobility. Frazzoli says they will advance the safety and comfort of automotives" and open the doors of mobility for people who cannot or should not drive; as he puts it, "if you had too much to drink, maybe you should let the computer take you home." Future autonomous vehicles can also increase the efficiency and throughput of our existing road system and help reduce congestion by coordinating with others cars. The autonomous vehicle will also be a "green vehicle" that can make more fuel"efficient decisions than human drivers. Future autonomous vehicles might save up to 20 to 50% of emissions and fuel consumption by optimizing speed and stopping.
In some ways, the autonomous vehicle is already with us. There are robotic components in the cars we drive today, like ABS (advanced breaking systems) which takes over control from human drivers to stop cars and avoid collisions. Frazzoli says that when ABS first reached the market it was a novelty and good drivers felt they could stop faster than the system. Today, it is widely accepted that it performs better than almost everyone except a Formula 401 racer. More recently, autonomous components enable cars to park themselves, and in the arena of mass transportation, there are autonomous bus demonstrations, and driver"less shuttles. Frazzoli sees the demand for autonomous vehicles as only growing. The autonomous vehicle can address the problem that both drivers and passengers are becoming increasingly distracted on the road with cell phones and electronics. And, with a growing network of V" 2" V (vehicle to vehicle) communications, autonomous vehicles can coordinate with other cars, beyond the capabilities of individual drivers. Simply put, human drivers can only account for vehicles they can see, while an autonomous vehicle has the advantage of synchronizing with the entire network upstream and downstream.
Some initial work on autonomous vehicles began in the 1980's. More recently it has captured the imagination of the US Defense Department whose DARPA program has set million dollar prize challenges. Frazzoli gives some firsthand lessons from participating in the highly coveted DARPA vehicle challenge of 2007. Initially there were 89 teams but only 7 teams, including MIT, made the final cut. The road conditions were very realistic and each vehicle team had to navigate as an urban vehicle in regular traffic. In other words, the autonomous vehicle needed to abide by the rules of the road, obey speed limits, merge into traffic, pass safely and avoid obstructions. Frazzoli jokes," the vehicle had to have enough skills to qualify for a California drivers license". The MIT team equipped a Land Rover LR3 with state" of _the" art" technology that included more than 40 CPUs, radar, and a myriad of laser like sensors, rotating scanners, and video cameras. Two main modules were a perception sub system, which figured out where the road was, and a planning control system, which then used a decision tree algorithm to anticipate and adhere to rules" of" the" road. Only six vehicle teams (including MIT) actually finished the race. According to Frazzoli, "the race was so successful that the only way you could detect that an autonomous vehicle was driving, was that each robotic driven car had a human driver close behind it with a remote control (in hand)."
About the Speaker(s): Emilio Frazzoli's main research interests are in the general area of planning and control for mobile cyber"physical systems, with a particular emphasis on autonomous vehicles, mobile robotics, and transportation networks.
He received a Laurea degree in Aerospace Engineering from the University of Rome, "Sapienza", Italy, in 1994, and a Ph. D. in Navigation and Control Systems from MIT Aero/Astro in 2001. Between 1994 and 1997 he worked as an officer in the Italian Navy, and as a spacecraft dynamics specialist for the European Space Agency Operations Centre (ESOC) in Darmstadt, Germany, and Telespazio, in Rome, Italy. From 2001 to 2004 he was an Assistant Professor of Aerospace Engineering at the University of Illinois at Urbana"Champaign. From 2004 to 2006 he was an Assistant Professor of Mechanical and Aerospace Engineering at the University of California, Los Angeles. He was the recipient of a NSF CAREER award in 2002. He is an Associate Fellow of the American Institute of Aeronautics and Astronautics and a Senior Member of the Institute for Electrical and Electronics Engineers. He is currently serving as an Associate Editor for the AIAA Journal of Guidance, Control, and Dynamics.
Associate Professor of Aeronautics and Astronautics
Laboratory for Information and Decision Systems at MIT.
Host(s): School of Engineering, Transportation@MIT
MIT World -- special events and lectures