backtop


Print 6 comment(s) - last by Coldfriction.. on Feb 20 at 6:02 PM


Nissan ESFlow EV Concept
It also wants to research Human Machine Interface for both technologies

Nissan has opened a new research center in Silicon Valley, where the automaker plans to focus on autonomous and connected vehicles.

The new facility is called the Nissan Research Center Silicon Valley (NRC-SV), and it will use partnerships with educational institutions and companies to work on projects for new vehicle technologies.

The initial projects in NRC-SV will be research of autonomous vehicles, research of connected vehicles that can use the Internet and infrastructure to communicate with the outside world, and research of Human Machine Interface to make both autonomous and connected vehicles more efficient.

“As we continue to expand our R&D capabilities throughout the world, we aggressively pursue our activities to create new values of mobility, by harnessing the latest information and communication technologies here in our new research center in Silicon Valley," said Nissan Executive Vice President supervising Research and Development Mitsuhiko Yamashita.

There were hints of Nissan's autonomous future back in October 2012, when the automaker announced that it would use steer-by-wire technology for future Infiniti models.

Source: Nissan



Comments     Threshold


This article is over a month old, voting and posting comments is disabled

How this will happen.
By Coldfriction on 2/20/2013 12:56:41 AM , Rating: 2
Let me explain how this needs to occur to happen. The system needs to be broken up into several fundamental pieces.

The first piece is local area position far more accurate than GPS. Transmitters behaving just like GPS need to be located along transit routes. These need to have the ability to function exactly like GPS while also correcting GPS postioning errors.

Second, each and every vehicle on the road must be mandated to contain a positioning system that uses the local position system. A vehicle must know within a tight tolerance of error where it is at all times. It must be able to communicate it's position, velocity, acceleration, physical limits, etc. to every other vehicle nearby. How do you mandate this? No different than mandating that cars have brake lights to communicte deceleration to other drivers.

Third, the infrastructure must be in control. The idea that a vehicle must be autonomous is an example of nerds seeking the coolest most complicated solution to a problem. A vehicle does not need to be all interpreting of its surroundings in an attempt to replace an intelligent human being. Each and every vehicle must communicate to this infrastructure and the infrastructure must be able to communicate back in the case of automated vehicles. This has to be true of all vehicles automated or not.

The concept is that the infrastructure establishes a virtual road on top of the physical road that an automated vehicle would be able to follow. Imagine virtural rails established along which an automate vehicle would track. The virtual road contains details that an an fully autonomous vehicle has a hard time interpreting. The infrastructure can tell exactly how fast a vehicle should be taking a radius based on the known coefficient of friction, super elevation, etc. of the road. In my opinion any system that requires wires or some other indicator embedded in the superstructure is a lost cause; virtual rails are the only way to go.

How do you implement this? Step 1: Standardize via the FCC some standard for local positioning systems. Step 2: Implement the LPS and network infrastructure along various roads, beginning with interstates. Step 3: Require all existing vehicles to be fitted with a LPS communication device (perhaps the most politically reprehensible part of the process). Step 4: Record the paths of existing manual vehicles and their behaviors to establish the virtual road system so that it corresponds with what people actually do. Step 5: Allow automakers to use the established standards to simply make vehicles that will follow the rails while avoiding other vehicles. This shouldn't be so hard as it is made out to be with the massive instrumentation of autonomous vehicles. A car only has a few inputs that control it. The ability to drive your car on non-automated roads and then allow it to drive itself when desired should be the next big thing in the automotive industry. Step 6: Continue automating all the major roads in the nation and trickle down to local road systems. Eventually having a system of virtual roads so complete that purchasing a car that does not require manual driving is feasible.

Benefits: Reduced deaths due to traffic accidents. Construction crews can turn off a section of road that shouldn't be traveled during construction on the fly instead of coning it off. Drinkers can get around without killing people. Underage driving becomes a non-issue and people could theoretically allow their children to have an increased mobility. Cars could be self locating where a family could more easily share one vehicle. Dad could go to work and then send the car home for his wife to use throughout the day. Every parking experience could be a valet situation and parking could be way out of the way in an unseen location. Long distance drivers could sleep and travel through the night. When you pull up to an intersection your car could tell you when to turn instead of some $200,000 intersection lights. A properly balanced system with only automated cars would be able to eliminate stopping at intersections. Fuels savings for convoys and the lack of stop and go traffic would be enormous. Traffic jams should be severly reduced or eliminated if enough vehicles are automated (think of 3 miles of stopped bumper to bumber vehicles accelerating simultaneously instead of in a wave pattern).

In economics this system would allow proper network balancing and usage of roads. A system could be put in place charging tolls based on usage instead of taxing all fuel and throwing that money into a pot (works for any kind of vehicle whether electric, fossil fuel, nuclear, etc.). A quick route in high demand should charge more money and a longer route in low demand should be cheaper to travel. Privatize roads and allow people to charge with a profit motive to set prices where they should be; I don't trust government to set prices such that free market principles will prevail. Using weighted paths traffic becomes a software optimization problem and not a human behavioral problem. You could pay more for the priveledge of getting where you want faster. Knowing which roads to build and improve would be clear as day and not subject to some board of transporation's decision with interests other than optimization in mind.

I say we repurpose some chunk of NASA and make this happen in the United States. They can send a pretty neat bit of tech to Mars but can't do something so simple as this that would change humanity forever. If you think the interstate system boomed the economy after it was built, you should realize that this type of system would do the same.

End super rant. DOT Engineer here frustrated that the coolest implemented advancements in transportation infrastructure are slower and far lamer than what is possible with todays technology. BTW if anyone wants to hire me to work on such a system as I have described I'm all ears.




RE: How this will happen.
By mjv.theory on 2/20/2013 5:03:27 AM , Rating: 2
quote:
I don't trust government to set prices such that free market principles will prevail.

Not sure I trust government decision makers to make all correct decisions either. But tell me, what is a "free market"? and has one ever existed?


RE: How this will happen.
By Coldfriction on 2/20/2013 6:02:17 PM , Rating: 2
When I say free market principles I mean that price needs to be set as high as the market will bear without a false cieling restricting it because it's a "right" or some such nonsense. A road is a scarce resource that should be priced high enough as to not overload its use. The high price allows funds to be allocated to the expansion and construction of more roads where the demand desires them.

There is a significant number of people that want price and cost to be the same thing. They aren't and shouldn't be. Price communicates demand such that those able to supply have some motivation to commit the effort to do so. Cost must be some measurable difference less than price. Profit is the difference between the two that provides the incentives that allow those with abilities and assets to expend them for the betterment of the world.

Concerning free markets in general, I say they are more common than most people realize. Who regulates Craigslist? Do you pay sales tax on used goods or something you buy from your buddy? Does the government stop you from trading a television for a gun because you didn't do it through the proper channels? Do you trade your time (cost to you) for a wage (price of your time) and make a profit? Most unhappy workers are the ones that feel they the trade doesn't have a high enough profit margin.

Nobody governs an ecosystem; its constraints are natural. I wish people would use the term "Natural Market" instead of "Free Market" as that would lend realization that the true natural laws of economics are as immutable as the true natural laws of physics. The only markets that have problems are the markets that aren't free.

Roads run by a government without prices and costs being determined by a free market make them more expensive and/or less useful than they could otherwise be.


"Folks that want porn can buy an Android phone." -- Steve Jobs














botimage
Copyright 2014 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki