The availability of several Advanced Driver Assistance Systems has put a correspondingly large number of inexpensive, yet capable sensors on production vehicles. By combining this reality with expertise from the DARPA Grand and Urban Challenges in building autonomous driving platforms, we were able to design and develop an Autonomous Valet Parking (AVP) system on a 2006 Volkwagen Passat Wagon TDI using automotive grade sensors. AVP provides the driver with both convenience and safety benefits - the driver can leave the vehicle at the entrance of a parking garage, allowing the vehicle to navigate the structure, find a spot, and park itself. By leveraging existing software modules from the DARPA Urban Challenge, our efforts focused on developing a parking spot detector, a localization system that did not use GPS, and a back-in parking planner. This paper focuses on describing the design and development of the last two modules.
[1]
Dirk Langer,et al.
Junior 3: A test platform for Advanced Driver Assistance Systems
,
2010,
2010 IEEE Intelligent Vehicles Symposium.
[2]
Hobart R. Everett,et al.
Where am I?" sensors and methods for mobile robot positioning
,
1996
.
[3]
Roman Henze,et al.
Reproducible transverse dynamics vehicle evaluation in the double lane change
,
2008
.
[4]
Sebastian Thrun,et al.
Junior: The Stanford entry in the Urban Challenge
,
2008,
J. Field Robotics.
[5]
Reid G. Simmons,et al.
A task description language for robot control
,
1998,
Proceedings. 1998 IEEE/RSJ International Conference on Intelligent Robots and Systems. Innovations in Theory, Practice and Applications (Cat. No.98CH36190).
[6]
Sebastian Thrun,et al.
Winning the DARPA Grand Challenge with an AI Robot
,
2006,
AAAI.