Scientists have developed a self-driving scooter, using the same technology that powers autonomous cars, which could help mobility-impaired people move around even in indoor spaces.
The researchers had previously used the same sensor configuration and software in trials of autonomous cars and golf carts, so the new trial completes the demonstration of a comprehensive autonomous mobility system.
A mobility-impaired user could use a scooter to get down the hall and through the lobby of an apartment building, take a golf cart across the building's parking lot, and pick up an autonomous car on the public roads.
The new trial establishes that the researchers' control algorithms work indoors as well as out.
"We were testing them in tighter spaces," said Scott Pendleton, a graduate student at the National University of Singapore (NUS).
The system, designed by researchers from Massachusetts Institute of Technology (MIT) and the Singapore-MIT Alliance for Research and Technology (SMART), includes several layers of software.
Low-level control algorithms enable the vehicle to respond immediately to changes in its environment, such as a pedestrian darting across its path. Localisation algorithms can be used to determine the vehicle's location on a map.
Map-building algorithms is used to construct the map, a scheduling algorithm allocates fleet resources and an online booking system that allows users to schedule rides.
Using the same control algorithms for all types of vehicles - scooters, golf carts, and city cars - has several advantages. One is that it becomes much more practical to perform reliable analyses of the system's overall performance.
"If you have a uniform system where all the algorithms are the same, the complexity is much lower than if you have a heterogeneous system where each vehicle does something different," said Daniela Rus, professor at MIT
"That's useful for verifying that this multilayer complexity is correct," Rus said.
With software uniformity, information that one vehicle acquires can easily be transferred to another. For instance, the scooter was tested in Singapore, where it used maps created by the autonomous golf cart.
Similarly, in ongoing work the researchers are equipping their vehicles with machine-learning systems, so that interactions with the environment will improve the performance of their navigation and control algorithms.
"Once you have a better driver, you can easily transplant that to another vehicle," said Marcelo Ang, an associate professor at NUS.
Software uniformity means that the scheduling algorithm has more flexibility in its allocation of system resources. If an autonomous golf cart is not available to take a user across a public park, a scooter could fill in; if a city car is not available for a short trip on back roads, a golf cart might be.
