Self-driving golf carts were successfully used to drive some 500 tourists around a garden in Singapore. The six-day experiment tested the carts skills when dealing with pedestrians and bicyclists, and involved an online booking system for scheduling pickups and drop-offs at 10 stations located around the garden.
“We would like to use robot cars to make transportation available to everyone,” said Daniela Rus, a senior author on the paper detailing the project, which was a collaboration between MIT and the National Research Foundation of Singapore. “The idea is, if you need a ride, you make a booking, maybe using your smartphone or maybe on the Internet, and the car just comes.”
The vast majority of those who used the carts were happy with the experience and said they’d use them again. In fact, some 95 percent of survey respondents said they would be more likely to visit the gardens again if the golf carts were a permanent feature.
“We are taking a minimalist solution to the self-driving-car problem,” Rus said. “The vehicles are instrumented, but they are not as heavily instrumented as the DARPA vehicles were, nor as heavily instrumented as, say, the Google car. 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.”The golf carts’ sensors use only off-the-shelf laser rangefinders, in fact, and a camera. The carts avoid obstalces by checking a cylindrical area surrounding the vehicle’s planned trajectory. The width and length of the cylinder change with the velocity of the cart. If an obstacle is detected within the cylinder, the cart’s computer plans a route to avoid it, by changing its trajectory, reducing the velocity, or some combination of the two.
Unlike cars driving an a standard road, the golf carts have an advantage: their relatively slow speeds give the computers more time to sort out new paths when an obstacle is encountered. But Rus says that slow-moving vehicles could have wider applications than merely gardens and golf courses.
“If you think about who needs rides,” she says, “it’s fast enough for the elderly population who no longer have a driver’s license and live in special areas where maybe their friend lives a mile away, and that’s too far to walk. If they want to go to the doctor or shopping, they can use the self-driving golf carts because that’s within some comfortable distance.”
Only one problem was encountered during the experiment: large, slow moving lizards. “It was this stop-and-go game over who’s going to do what,” Rus says.