Nissan: Putting lessons from Mars to work on Earth – insights on unconstrained car investigate from Nissan’s Dr. Maarten Sierhuis

The Nissan Research Center (NRC) attracts tellurian talent as a core for open creation by pity a appealing qualities and providing information to a public. In a letter below, a NRC presents insights on unconstrained automobile investigate and synthetic comprehension from Dr. Maarten Sierhuis, a executive of a NRC in Silicon Valley, who is regulating his knowledge during NASA to work on unconstrained expostulate record for Nissan.

Sierhuis came to a U.S. in 1989 from a Netherlands and worked for IBM and NYNEX Science and Technology until 1988. After earning his PhD in synthetic comprehension during a University of Amsterdam, he worked for NASA and Xerox PARC. He spent 12 years with NASA where he grown a mechanism denunciation underpinning intelligent systems for use in robots, spacesuits and NASA’s Mission Control Center. In 2013, he jounced Nissan where he heads a Nissan Research Center in Silicon Valley and leads mixed teams of researchers operative on unconstrained vehicles, connected vehicles and human-machine communication and interfaces.

A Career that Began in Space

I started operative during NASA Ames Research Center in Silicon Valley in 1998. After 12 years there, we went to Xerox PARC, where we served as executive and ran investigate on multi-agent systems and human-machine interaction. It was during NASA, though, that we combined many of what I’m putting to work for Nissan today. We started with growth of a make-believe denunciation that authorised us to indication tellurian function and mixed people operative together. We were looking during how people competence live on Mars and work with people behind on Earth, as good as unconstrained systems, including robots and intelligent habitats on Mars. We started initial simulating this with a language, though once we started using this denunciation in genuine time it also became a programming denunciation for unconstrained systems in general.

We started building intelligent agents for robots, for Mission Control and for a habitat. Then we combined a speech-dialogue complement to this — now we could have astronauts articulate to their unconstrained system, including a drudge and a habitat, as good as to systems in Mission Control. We put it to work in space; we indeed built an intelligent representative in a space fit to guard a astronaut’s health autonomously.

We helped pattern how robots would work on Mars with people on Earth. After this success, we were asked to automate moody controllers in NASA’s Mission Control Center for a International Space Station. The final plan we did during NASA used my mechanism denunciation to automate a moody controller for a ISS. This complement went live in 2008 and it’s still in use, with all communication from a ISS holding place by it.

The Connection with Cars


To build a unconstrained complement for a automobile on Earth is unequivocally like building a drudge that drives 80 miles an hour unequivocally tighten to other robots. That is unequivocally opposite from Mars, where there are not that many people – during slightest not yet! Many issues come adult when we consider about humans interacting with any other and with robots, since a automobile needs to be on a highway with other people — pedestrians, bicyclists and other cars. The thought of multi-agent displaying becomes key: meaningful what everybody is doing, so that a automobile knows not usually what it needs to do itself, though also a attribute with others on a road. In civic areas, we have to understanding with pedestrians, bicyclists, motorcyclists, cars, animals — a whole spectrum of communication becomes a unequivocally critical study. The work that we did during NASA is unequivocally applicable in this context.

In tie with these pushing sourroundings issues, we started a investigate plan in North Holland, a Dutch range with a world-leading trade government system. Every trade light there is connected with all a rest, and they promulgate boldly formed on how bustling a roads are to confirm when they will change. We used information from these trade lights in intersections and built machine-learning algorithms to envision when a trade light will go red or green, formed on how distant one is from a light.

We now use this algorithm with a unconstrained automobile software, so as a automobile drives it gets information from a trade complement to envision how prolonged a light will be red or green. Ideally, a unconstrained automobile doesn’t have to stop; it can automatically reroute and take lights that are green. As we optimize a unconstrained complement to equivocate interlude for lights, we also send information from a automobile behind to a complement so as to optimize trade government on a incomparable scale.

Why Silicon Valley

In unconstrained vehicles now, many of a record is formed on program and on synthetic comprehension (AI). There’s no improved place to do this work than in Silicon Valley. The NASA Ames Research Center was one of a initial places where robots and unconstrained systems were put together, and all a record around unconstrained vehicles was grown during universities and companies in a area. In a early 2000s, many AI researchers from around a universe came to Silicon Valley to join a IT industry. Nissan satisfied this in a mid-2000s: If we wanted to be critical about building unconstrained automobile record in house, we had to have a participation in Silicon Valley.

The Nissan Research Center has relations with people from Stanford and UC Berkeley — that yield a pivotal talent pool for us, too — and we’ve set adult unequivocally close, fruitful investigate partnership with NASA, usually down a highway from us. We’re vocalization with a series of Silicon Valley firms to see how we can work together. So it’s critical to be in this region.

Looking Toward a Future of Mobility

At NASA, we researched how humans and robots would work together on Mars in a future. When Nissan asked me to do that for vehicles on Earth, it was unequivocally formidable to contend no. we have a sold perspective on how humans and unconstrained systems should work together, and we unequivocally appreciated that people during Nissan had a identical idea. Nissan believes that mobility is for a good of society; that’s one of a reasons we motionless to come here.

It’s sparkling to consider about a multitude where a right mobility complement is used for a right purpose during a right time. We’re going to have trains interacting with common vehicles that can seamlessly take me from work to home to propagandize to collect adult my children. Whatever we need to do in my life will be seamlessly integrated with mobility services during opposite places. we don’t trust that open travel should go away. By integrating a vehicles – a trains, a planes, even a bicycles together into a multitude where we have some-more space for parks and pleasing landscapes – we’ll have some-more space for people.

Nissan’s prophesy of probably 0 fatalities and 0 emissions is unequivocally a good motivator for doing investigate to pierce in that direction. Autonomous record can be practical not usually inside a vehicle, though also in a cloud, in trains and in other travel systems. We should always be fit in a proceed we pierce around and correlate with others.

Based on an talk carried out in Sep 2016.


For years, #Nissan’s Maarten Sierhuis has worked with former colleagues during @NASA to rise a new proceed to unconstrained driving. #CES2017

Seamless Autonomous Mobility: The Ultimate Nissan Intelligent Integration


More from Nissan during CES 2017


# # #