SMART Driverless Vehicle Software (14415S-FM)
This invention enables suitably equipped automobiles to drive themselves from an origin to a destination point on a known road network while avoiding collisions with obstacles, pedestrians, or other vehicles.
The system relies on the availability of a number of sensors, either “external” (LIDARs, video cameras) or “internal” (wheel odometry, inertial sensors, and actuator position sensors). The system can handle simple road networks such as those found at the National University of Singapore. The software does not include the ability to navigate complex intersections, and/or to detect and recognize traffic signals.
The software’s capabilities have been tested and demonstrated on the autonomous vehicle fleet of the SMART-FM project at NUS, including golf carts and an electric city car based on the Mitsubishi iMiEV, called Shared Computer- Operated Transport (SCOT),
The main modules that have been implemented on the current system include perception, mapping-and-prediction, localization, planner and controller.
For more information on this technology contact: