Politics.be - Sensoren voor zelfrijdende auto’s?

Politics.be (https://forum.politics.be/index.php)

- Wetenschap & Technologie (https://forum.politics.be/forumdisplay.php?f=224)

- - Sensoren voor zelfrijdende auto’s? (https://forum.politics.be/showthread.php?t=241366)

Sensoren voor zelfrijdende auto’s?

Er zijn vele video’s te zien over “zelfrijdende auto’s” en toch heb ik een vraag.
Ik dacht dat er sensoren aangebracht moesten worden langs de route’s van zulke auto’s, om te kunnen navigeren.
Heb ik dat verkeerd gedacht en zijn zulke sensoren niet nodig? Doen ze dat handiger via satellieten?

Tesla heeft cameras op de wagen met beeldherkenning.

8 kwaliteitsvolle camera's en 1 radar en een hoop sonars, 12 stuk, rondom ... zouden voldoende zijn volgens Tesla/ SpaceX , en die hebben er zeker verstand van... ;-)

Natuurlijk veel rekenpower en de snelste GPU's (AI), maar daarvoor is nog wat tijd, ook voor upgrades moest het niet snel genoeg zijn.... Ook voor de software, die steeds beter zal worden. Opties 5G, V2V, V2X mogelijkheid zal ook wel voorzien zijn.

* ff ggl op enhanced autopilot hardware:

Citaat:

https://en.wikipedia.org/wiki/Tesla_...lot#Hardware_2

Hardware 2, included in all vehicles manufactured after October 2016, includes an Nvidia Drive PX 2[30] GPU for CUDA based GPGPU computation.[31][32] Tesla claimed that Hardware 2 provided the *necessary equipment to allow full self-driving capability at SAE Level 5. The hardware includes 8 surround cameras and 12 ultrasonic sensors, in addition to forward-facing radar with enhanced processing capabilities.[21] The Autopilot computer is replaceable to allow for future upgrades.[33] The radar is claimed to be able to observe beneath and ahead of the vehicle in front of the Tesla; the radar can see vehicles through heavy rain, fog or dust.[34]

Hardware specifications

2016 Enhanced Autopilot/Full Self-Driving Capability[35]
Hardware 2

- Radar: range 160 m (525 ft)

- Forward Cameras:
Trifocal camera:
Narrow: 250 m (820 ft)
Main: 150 m (490 ft)
Wide: 60 m (195 ft)

- Forward Looking Side Cameras:
Left: 80 m (260 ft)
Right: 80 m (260 ft)

- Rearward Looking Side Cameras:
Left: 100 m (330 ft)
Right: 100 m (330 ft)

- Rear View Camera 50 m (165 ft)

- Sonars: 12 surrounding with 8 m (26 ft) range

- Platform: NVIDIA DRIVE PX 2 AI computing platform[37]

Function specifications (pending incremental updates and regulatory approvals):
2016 Enhanced Autopilot/Full Self-Driving Capability Hardware 2

- Hands-on feature with limited Hands-Free On-Ramp to Off-Ramp for limited-access roads: Yes[39]

- TACC-Traffic-Aware Cruise Control (Smart/Adaptive Cruise Control): Yes[39]

- Max speed : 90 mph (145 km/h)[42]

- Autosteer: Yes, including tighter, more complex roads[43]

- Auto Lane Change: Automatically done all by itself for a faster lane and without driver's judgement nor input.[45]

- Highway Interchanges: Yes[45]

- Autopark: Parallel and Perpendicular Parking Yes[47]

- Summon (remote automatic car retrieval, including automatic garage door opening and closing): Yes[47]

- Lane Departure Warning: Yes[50]

- Full Self-Driving Capability: Yes, with an additional fee. The Tesla car will be able to drive itself, automatically recharge at "cable bot"-equipped Superchargers and can use Parking Seek to find a parking space all without a driver.[47]

https://en.wikipedia.org/wiki/Vehicle-to-vehicle
https://en.wikipedia.org/wiki/Vehicle-to-everything
https://en.wikipedia.org/wiki/5G

en...

Citaat:

https://venturebeat.com/2017/05/20/w...-cars-smarter/
Why Intel believes 5G wireless will make autonomous cars smarter
MAY 20, 2017

Citaat:

Oorspronkelijk geplaatst door ron kreike (Bericht 8583690)

Om op dit ogenblik de navigatie in uw auto te laten functioneren heb je satellieten nodig.

De toekomst van de satellietnavigatie bestaat erin om tot op een paar centimeter nauwkeurig de positie van een wagen te bepalen.

BRON
_{3 jan 2017}

De technologie hierachter:

Germany’s Bosch and Geo++, U-blox of Switzerland, and Japan’s Mitsubishi Electric announced the establishment of Sapcorda Services last Tuesday, a joint venture to provide global navigation satellite system (GNSS) positioning services of centimeter-level accuracy via
satellite transmission,

mobile cellular technology,

and the Internet.

BRON
_{20 aug 2017}

(fig.: de satellieten van Galileo halen een accuraatheid van 1 meter)

Accuraatheid tot op het centimeter-niveau.

Via GNSS = global navigation satellite system (GNSS) positioning services.

Sapcorda Plans Centimeter-Scale GPS for Europe
_{15 aug 2017}

Sapcorda Services is een joint-venture met als partners:

Bosch
Geo++ http://www.geopp.de/
U-blox https://www.u-blox.com/en
Mitsubishi Electric

Sapcorda http://www.sapcorda.com/

(...) Executives of the four companies said meeting the demands of centimeter-scale GNSS positioning required for applications such as autonomous driving necessitated the creation of real-time correction data services. The U.S.-operated Global Positioning System (GPS) is accurate only to about 10 meters, while Europe’s corresponding Galileo system, with 18 of its 24 satellites currently in orbit, has an accuracy of 1 meter. Yet even 1-meter accuracy is not sufficiently precise when it comes to ensuring that autonomous vehicles, for example, maintain safe lane positioning and avoid certain obstacles. (...)

Tesla gaat AI-chip ontwikkelen voor autonome voertuigen (Tesla zal wel de eerste zijn). Huurt daarvoor Jim Keller van AMD.
Indien de autopilot-hardware tegen dan niet voldoet wordt die gratis gewisseld:

Citaat:

https://electrek.co/2017/12/08/elon-...ip-jim-keller/

Elon Musk confirms that Tesla is working on its own new AI chip led by Jim Keller

Fred Lambert
- Dec. 8th 2017

There have been rumors that Tesla is developing its own chip optimized for self-driving applications ever since we reported on the automaker quietly hiring legendary chip architect Jim Keller from AMD last year.

Now CEO Elon Musk finally confirmed the rumor – specifying that Tesla is working on its own new AI chip and that the effort is led by Keller.

At the Conference on Neural Information Processing Systems (NIPS) yesterday, Musk held a fireside chat with Tesla’s new Director of AI and Autopilot Vision, Andrej Karpathy, Jim Keller, Tesla’s Vice-President of Autopilot Hardware Engineering, and Shivon Zilis, a partner at Bloomberg Beta and a project director for Musk’s office.

According to people at the event, the discussion revolved around Tesla’s artificial intelligence work and it was seemingly a recruiting effort for the company.

But Musk also made a few interesting comments, including a confirmation that Tesla is working on its own AI chip to enable self-driving technology.

The CEO confirmed that Keller is behind the effort, which he believes will result in “the best custom AI hardware in the world.”

Keller has an important reputation in the hardware world after having been a key player in the creation of several architectures, like Athlon K7 and K8 at AMD, before developing Apple’s A4 and A5 processors, which powered most of Apple’s mobile devices from 2010 to 2012. He later came back to AMD and led the design of the company’s latest chip architecture.

After quietly joining Tesla in late 2015, we reported that the automaker poached a team of chip architects and executives from AMD following Keller’s hire.

That’s when the speculation that Tesla is working on its own chip, presumably for self-driving, started. It was also in the middle of Tesla changing hardware from Mobileye to Nvidia for its Autopilot program.

Tesla is now up to its Autopilot ‘2.5’ hardware suite with more computing power for autonomous driving based on Nvidia’s Drive PX platform.

Musk previously said that he expects Tesla’s 2.0 and 2.5 Autopilot hardware suites to enable “fully self-driving”, but if it doesn’t, Tesla will replace the computers in those vehicles for free.

And now we know that Tesla is working on computers with its own AI chips.

When asked about potential improvements at the event yesterday, Keller said that there are opportunities to remove overheads between components for specialized hardware, which Musk says could result in 10 times the current power for a fraction of the cost.

Computing power inside vehicles is becoming increasingly important and like several other generally outsourced vehicle components, Tesla seems dead serious about bringing it in-house.

Sapcorda gaat een open formaat creëren.
Die openheid moet ervoor zorgen dat het systeem een standaard wordt.
Mitsubishi brengt z'n ervaring uit Japan mee.

The scheme is similar to Japan’s GPS-augmented Quasi-Zenith Satellite System (QZSS), which aims to provide centimeter-level positioning and navigation services in Japan. The system employs 1,300 reference stations to constantly correct satellite errors. The corrected data is then compressed from an overall 2-megabit-per-second data rate down to 2 kilobits per second for real-time transmission back to the satellite for broadcasting to user receivers. Shibata noted that Sapcorda-corrected data will be delivered in an open format to encourage system integrators, original equipment manufacturers, and terminal manufacturers to work with its specifications. “The hope is,” said Shibata, “that providing this open format will help create an industry standard.” BRON

(fig.: Japan’s centimeter-level navigations system relies on fleet of satellites in geostationary orbit)

Accuraatheid op centimeter-niveau met de hulp van satellieten is één van de voorwaarden voor de zelfrijdende auto.

Een tweede voorwaarde is dat er 3D-kaarten beschikbaar zijn van het wegennet. Wat doet men in Japan?

Go Takezawa, executive director for QZSS Strategy at Japan’s National Space Policy Secretariat, in updating the press last Tuesday, said that QZSS will help aid the creation of 3D maps, another essential component necessary to make driverless vehicles operationally safe. Currently, the Dynamic Map Platform Co. (DMPC), a government-backed joint venture between major Japanese auto manufacturers, Mitsubishi Electric, and map-related companies, is creating 3D maps of 30,000 kilometers of Japanese highways. The company is capitalized at 4 billion yen (US $36.4 million) a third of which the Japanese government is funding through the public-private Innovation Network Corp. of Japan, DMPC’s major investor.

Enkele andere toepassingen van GNSS, anders dan de zelfrijdende auto:

Other applications Takizawa cited include precision control
of autonomous tractors in agriculture,

of machinery used in large-scale construction,

and of drones employed in delivering goods to small islands close to Japan.

Takizawa noted that tests of all these applications are ongoing, and good results have been obtained in all cases.