If your idea of a personal motorized transportation is a Rascal electric scooter like the one George uses to try to escape from a similarly equipped elderly gang in Seinfeld, or from another George, President George W Bush, losing his balance on a Segway, consider yourself lucky. You can find humor in a dire situation for millions. The elderly and infirm who have no or limited ambulatory function are not laughing.
It is them that Tony Hong founded Strutt. Singapore, Hong’s native country, has the 3rd highest proportion of elderly people in the world. For them and others disabled in their early years, he recently presented the STRUTT ev1 at San Francisco’s Fort Mason. It was clearly a next-generation motorized wheelchair: It gets from one place to another as you and I would, by sensing its surroundings, not by following a route or being precisely guided by the person on it. More on that later.
In fact, ev1, with its LiDAR and AI, more closely resembles a Waymo in function if not in form. It’s like an autonomous vehicle for one, small and smart enough to move around the home.
From the Presentation
Tony didn’t set out to make a better wheelchair. He set out to rethink personal mobility. The result is perhaps the first personal mobility vehicle that can figure out on its own how to get from Point A to Point B — even in a cramped apartment.
Tony, along with his chief designer, Barney Mason, showed how the ev1 sees the world through LiDAR scans, with the point clouds projected on the big screen for the audience.
We see a short video featuring several individuals who spoke of the ev1 as something that would change their lives, even make life worth living again. I’ve sat through many a tech demo. This was the first that could have brought tears to my eyes.
The ev1 came across as not just an AI-powered vehicle bristling with sensors (it has plenty, including two LiDAR scanners, two more than a Tesla robotaxi) but a mobility solution built for and around people, one that could not only move them around, but also give them the confidence and spontaneity and a go-anywhere attitude that many wheelchair-bound people lack.
A Founder Uses Science to Help Needy Humans
Tony Hong grew up in Singapore with an engineer father and a nurse mother. The combination — technical rigor and human care — put him on his current path.
As a young student, Tony blazed through physics competitions, won the Presidential Research Award while at the Hong Kong University of Science and Technology (the first student ever to do so), and went on to UC Berkeley for a PhD in physics. He has been published in Science and Nature, and for 4 years, he was a professor at China’s Southern University.
While on the path to a career in academia, there was something nagging him.
“I kept asking myself, what is the real purpose of the work?” Tony said. “I saw amazing technologies in Silicon Valley — but many were disconnected from real human needs.”
If autonomous vehicles can navigate complex city streets, if robots can dance, if AI can plan flight trajectories — why can’t mobility devices help people travel a simple hallway without fear?
A Crisis Nobody Talks About
Countries across the world now have 20–30% of their population over age 65, while the number of people available or willing to help is shrinking. And mobility devices have changed little over the past century.
Technology should not be just for the future, says Tony, it should help people today.
Strutt Starts
Strutt started with meeting and talking to users of existing scooters and wheelchairs, who had spent years navigating sidewalks, markets, and homes with underpowered scooters or manual wheelchairs. They heard stories of frustration, exhaustion — a psychic pain that compounded their physical inconvenience. They were in a world in which they didn’t fit or move around in. They blamed themselves.
We see videos of them, victims of degenerative disease, the lost independence from always having to be helped, fear of falling over, growing unease of the outdoors, of being acutely aware of uneven sidewalks, or worse: no sidewalk. One user had a horse she couldn’t visit because her scooter couldn’t navigate the rough terrain.
“I don’t have to rely on my partner pushing me up the hill,” says Joe Motta. His San Francisco apartment is on a street with an 11°-degree incline.
Design Grounded in Reality, Not CAD
Barney Mason, Strutt’s head of design says the team didn’t begin with specifications. They began with listening.
One early adopter of the ev1, Jennifer, lives in a neighborhood with a speed bumps that she had learned to circumvent. Her motorized wheelchair couldn’t handle it. Upon the Strutt team’s coaxing, she tried the speed bump. The ev1 goes over the speed bump flawlessly.
When asked how the ev1 could be improved, Jennifer suggested more supportive armrests. The Strutt team not only changed the armrests, they took the opportunity to do a whole redesign — even though ev1was in the final design phase before production.
“Not a single component was left untouched,” says Barney.
The ev1 can be used indoors and out, but with a range of 20 miles, you can’t count on it to go across town. For that, the personal transporter must itself be transported. Strutt takes this into account.
The ev1is designed to be taken apart and reassembled. It’s modular, we hear. It breaks down into five lightweight pieces that can be lifted into a car trunk by an able-bodied caretaker or partner. The modules are rounded and smooth, with built-in handles. Should a van with a ramp be available, the ev1can back itself up the ramp on its own and dock.
The dashboard, for lack of a better term, is a touchscreen about the size of an iPhone.
How’s It on Cats?
Tony took the stage again, switching into engineering mode. We see point clouds generated by the twin LiDARs.
“This is what that ev1 is seeing onstage,” said Tony, pointing to the screen as the ev1 moves around the room. Even with the low resolution of the point cloud, we could make out obstacles, people nearby and very importantly, objects at floor level.
In a home environment, it’s important to recognize objects on the floor, like pets and feet, Tony explains as he sticks out his foot in front of the ev1, now advancing towards him. The ev1 stops well before his foot.
We don’t see any demonstration for pets, but in light of a recent incident in which a Waymo crushed a cat, we can’t help but ask:
“So the ev1 won’t be running over any cats?”
No, says Tony, laughing.
Sensors that Make Sense
EVSense is Strut’s perception and intelligence platform, combining:
- Two automotive-grade LiDARs
- Ten depth sensors (ToF)
- Ultrasonic sensors for transparent surfaces (glass walls, sliding doors)
- Multiple cameras for semantic understanding
- An onboard AI compute module capable of local, real-time perception and trajectory planning
It is, Tony says, “the most advanced sensor stack ever used in a personal mobility device.”
Why so much technology? Isn’t a joystick control good enough? Controlling a personal mobility vehicle is actually harder than driving a car, Tony explains.
A car’s steering wheel has a 36:1 ratio — a huge amount of error tolerance for human control. A mobility joystick is effectively a 1:1 movement. In other words, it’s touchy. A slight touch can make it lurch ahead or whip the vehicle around. Users are required to have or develop a surgeon-level precision in order to negotiate their vehicles.
The ev1 changes all that by making the controls more forgiving.
The Mind of the Machine: CoPilot, CoPilot+, and Pathfinder
The ev1 doesn’t “auto-drive” in the sense of a robotaxi. It’s more like an airplane being flown on autopilot.
With an aviation autopilot, the human pilot still directs the aircraft but does it at a macro level, adjusting for conditions, avoiding collisions.
The ev1operates the same way. The user tells it where to go and it figures out how to get there, adjusting for terrain, going around obstacles.
The CoPilot prevents collisions. Simple as that. It’s when the environment is tight that this is most important.
A CoPilot+ mode adds real-time steering assistance. The user gives it macro commands with the joystick, like pushing it forward. The ev1 determines the safest, smoothest actual path.
On the screen, Tony shows the difference with a red line that signifies the user intent and a blue line that the ev1 determines to be the safe and smooth path.
All the time an ev1is being used in a home, it is learning. With repeated passes giving it more and more data points, the home acquires higher and higher resolution. At any time, the user can tap a point on the screen and ev1will go there without further input.
Users can also interact with the ev1 to create waypoints. For example, a user can say: “I am at the fridge,” and the ev1will mark the spot for future reference. It is smart enough to understand that when you say “I want some milk,” it needs to take you to the fridge.
The ev1 also has a Follow Mode by which a person, such as a caregiver, can walk along, effectively teaching the machine a path.
Tony hints at a future with an even more advanced interface, a brain control interface (BCI).
Quadra Drive System
The Quadra drive system of the ev1 consists of:
- Active steering on the front wheels, no drive motors
- Two rear wheels with drive motors, each capable of 120 Nm of torque.
That’s comparable to that of a small city car, says Tony.
The suspension was revamped. With swappable springs, the ev1can manage a rider of up to 350 pounds.
How’s It On Hills?
Joe Motta, who has been testing the ev1 inside and out, once had an extremely active life. He would like to say he is now wheelchair bound because of something brave and daring, like a parachuting, but it was multiple sclerosis that laid him low. He was positively delighted to find the ev1 could go up a steep (11°) hill in his San Francisco neighborhood. hills with confidence.
The ev1 can also traverse trolley tracks used by the city’s famous trolley cars.
The ev1 is able to cover up to 3-inch wide gaps, says Tony, making it get safely cross train/platform gaps.
The ev1 is shown being able to turn the corner to the right even while encountering a ramp to the left — effectively creating a negative banking angle that would capsize a conventional 4-wheel scooter. I wasn’t clear how the ev1 managed this so I asked Tony after the presentation.
It can “see” the ramp, Tony explains, and will slow down in anticipation.
A Futuristic Mobility Platform
The ev1 is currently in final testing, says Tony.
What are you waiting for? I ask.
“We want to make sure we get everything right,” he says.
In the meantime, the ev1 has been recognized by CES judges in both Innovation and Design categories. [Full disclosure: author is a CES judge in the Innovation category.]
The ev1 is what happens when advanced robotics, autonomous-vehicle engineering, and human-centered design collide with empathy.
Personal human mobility deserves as much, if not more, consideration than we give to autonomous vehicles. The stakes are much higher for personal mobility. A robotaxi can give you a ride across town, and in so doing, equal to and replaces a human driver. It is only an incremental improvement over a human driver at best and is normally enjoyed by the affluent. By contrast, an intelligent personal autonomous vehicle is a marked improvement for the neediest among us.