Velodyne LiDAR and SMART have collaborated on the experimental deployment of fully autonomous golf carts, guided by technology that measures distance by illuminating a target with a laser light. In the experiment, self-driving golf carts ferried 500 tourists around winding paths trafficked by pedestrians, bicyclists, and the occasional monitor lizard.

Velodyne LiDAR has collaborated with SMART, the Singapore-MIT Alliance for Research and Technology, on an experimental deployment of fully autonomous golf carts, guided by technology that measures distance by illuminating a target with a laser light.

SMART conducted “an experiment over six days at a large public garden in Singapore, in which self-driving golf carts ferried 500 tourists around winding paths trafficked by pedestrians, bicyclists, and the occasional monitor lizard.” SMART is a partnership between MIT and the National Research Foundation of Singapore.

For the experiment, researchers from MIT and Singaporean universities enlisted two modified Yamaha electric golf carts, each equipped with a webcam and Velodyne’s 16-channel real-time 3D VLP-16 LiDAR Puck, which replaced earlier 2D sensors. The carts transported passengers along the paths while autonomously navigating and watching for such obstacles as pedestrians and animals. The golf carts jockeyed for position along with everyone, and everything, else on the paths.

Carts cruised at a top speed of about 15 miles per hour, relying on a “dynamic virtual bumper”—a cylinder surrounding the vehicle’s planned trajectory—rather than GPS. The cylinder’s width and length changed in response to the vehicle’s velocity. When an obstacle entered the cylinder, the vehicle’s onboard computer modified the cylinder to exclude it, altering the trajectory, reducing the velocity, or both.

The experiment also tested an online booking system through which visitors could schedule pickups and drop-offs at any of 10 stations around the garden, routing and redeploying the vehicles automatically to accommodate requests. When surveyed about the experience, 98% of passengers said they would like to ride the carts again and 95% would return to the garden if the carts became a permanent attraction.

“We [took] a minimalist solution to the self-driving-car problem,” said Daniela Rus, a professor in MIT’s Department of Electrical Engineering and Computer Science. “We believe that if you have a simple suite of strategically placed sensors and augment that with reliable algorithms, you will get robust results that require less computation and have less of a chance to get confused by ‘fusing sensors,’ or situations where one sensor says one thing and another sensor says something different.”