A driver rides hands-free in a Tesla Motors Inc. Model S vehicle equipped with Autopilot hardware and software in New York, 2016. At the end of August the Israeli startup company Oryx Vision, which had been developing a light detection and ranging camera system (LiDAR) for autonomous cars, announced it was shutting down. The company had raised $67 million in capital and was returning a remaining $40 million to investors.
Orxy Vision ’s founder, Ran Wellingstein, said that market conditions had changed and that the time it would take for the first mainstream autonomous vehicles to reach the market would take longer than he and others had anticipated.
“This is a game for players that have the patience to wait five or 10 years, so it’s a problematic situation for backing by venture capital,” he told TheMarker at the time.
The startup industry that’s been created around the autonomous car was based on the assumption that a revolution was just around the corner, All young firms needed to do was to develop the support systems, mainly a system of sensors of computer vision.
But more recently, the feeling has developed that technological and regulatory obstacles are going to take a lot longer time to clear than they had predicted.
“The industry believed that a completely autonomous car would be there by 2023, but today we realize that goal is more distant and that realization has resulted in a major change in the automobile industry,” said Yakir Machluf, who leads the mobility team at the Israel-based venture capital fund OurCrowd.
Oryx Vision was the first Israeli causality of the new forecast, but it wasn’t the first globally. Last June, the U.S. startup drive.ai once valued at $200 million announced it was ending operations to develop autonomous cars – only to be saved at the last minute by being acquired by Apple in order to absorb its top staff.
It’s seems certain other shutdowns will follow suit, the only question being who will survive. To that, Machluf answers: “Companies that adapt themselves to lower levels of autonomous-vehicle technology.”
The U.S. The National Highway Traffic Safety Administration defines five levels of autonomous cars. At SAE (for Society of Automotive Engineers) Level 1 an automated system can sometimes help the human driver with some parts of driving. At Level 2, the system can conduct some parts of the driving task, while the human driver monitors the driving environment.
At Level 3, the system can conduct some parts of the driving task and monitor the environment in some instances, but the human driver must be ready to take back control. Levels 4 and 5 are fully autonomous vehicles (although in the case of 4 only under certain conditions).
The latest estimates are that vehicles at Level 4 will only reach the market in 2026. OurCrowd , which has stakes in autonomous-vehicle startups, is now readying itself for the delayed new world.
“The autonomous vehicle is now further away because the task is very complicated. Today we understand that the quality of sensors and their accessibility have to be better,” said Machluf. “Everything has to work even when there’s rain coming down, so at the end of the day we’ll need a very exact laser system and phenomenal resolution as well as radar that can work in bad weather and multidimensional cameras.”
A Level 5 autonomous car will need elements of everything, he said. Meanwhile, some of the existing technology can be added into cars as a driver-assistance system that can help save lives.
Machluf is thinking of startups like Tel Aviv-based Arbe Robotics, which is one of OurCrowd’s portfolio companies and developing 4D-image car radar. Machluf said Arbe’s technology can be adapted for Level 2 functionality and that’s why some of the world’s leading automakers are showing an interest in it.
OurCrowd is among the biggest venture capital funds in Israel, according to the industry monitor Pitchbook. OurCrowd has invested so far in 16 auto-tech companies, making it Israel’s second-biggest player in the field after the specialist fund Maniv Mobility.
Eli Nir, a general partner and head of investments at OurCrowd, talks about a portfolio company called Ride Vision that is developing a driver-assistance technology similar to what Mobileye has done, for motorcycles.
“Motorcyclists are hurt a lot in road accidents, but a big share of the accidents are caused by the rider, Ride Vision is in a super-interesting niche because ADAS [advanced driver-assistance systems] for two-wheeled vehicles is completely different from one for a car,” Nir said, quoting Mobile founder Prof. Amnon Shashua saying his company wouldn’t touch the segment because of the differences in technology requirements.
Another portfolio company he points to is VayaVision, which gathers all the information collected by a vehicle’s sensors (radar, LiDAR and cameras) and forms a single picture from it.
But OurCrowd isn’t investing just in autonomous vehicles. “There is a market consensus today that the minute autonomous cars reach the roads, there will be a 5-7% increase in traffic. Today, we believe it will be even more because autonomous cars will give mobility to people who don’t have a license or those who can’t drive, like children,” said Nir.
OurCrowd’s partners think the future of transportation lies in integrating different transportation solutions—private cars, ride sharing, public transportation and light transportation, like electric scooters that can supply the last leg of a trip for many. OurCrowd had invested in one such company, called Jump, that was sold to Uber in April 2018 for $200 million.
Now it has a stake in another firm called Superpedestrian, which is developing the next-generation scoote.
“This is a company directed at manufacturers of high-quality scooters, because today scooters are off-the-shelf products that are inappropriate for shared transportation and last for just two or three months. It’s not long enough [for operators]to make a profit,” said Nir.