Nvidia Drive

Nvidia Drive is a computer platform by Nvidia, aimed at providing autonomous car and driver assistance functionality powered by deep learning.[1][2] The platform was introduced at the Consumer Electronics Show (CES) in Las Vegas in January 2015.[3] An enhanced version, the Drive PX 2 was introduced at CES a year later, in January 2016.[4]

Maxwell based

The first of Nvidia's autonomous chips was announced at CES 2015, based on the Maxwell GPU microarchitecture.[5] The line-up consisted of two platforms:

Drive CX

The Drive CX was based on a single Tegra X1 SoC (System on a Chip) and was marketed as a digital cockpit computer, providing a rich dashboard, navigation and multimedia experience. Early Nvidia press releases reported that the Drive CX board will be capable of carrying either a Tegra K1 or a Tegra X1.[6]

Drive PX
Drive PX

The first version of Drive PX is based on two Tegra X1 SoCs, and was an initial development platform targeted at (semi-)autonomous driving cars.

Pascal based

Drive PX platforms based on the Pascal GPU microarchitecture were first announced at CES 2016.[7] This time only a new version of Drive PX was announced, but in multiple configurations.

Drive PX 2

The Nvidia Drive PX 2 is based on one or two Tegra X2 SoCs where each SoC contains 2 Denver cores, 4 ARM A57 cores and a GPU from the Pascal generation.[8] There are two real world board configurations:

  • for AutoCruise: 1× Tegra X2 + 1 Pascal GPU
  • for AutoChauffeur: 2× Tegra X2 + 2 Pascal GPU's

There is further the proposal from Nvidia for fully autonomous driving by means of combining multiple items of the AutoChauffeur board variant and connecting these boards using e.g. UART, CAN, LIN, FlexRay, USB, 1 Gbit Ethernet or 10 Gbit Ethernet. For any derived custom PCB design the option of linking the Tegra X2 Processors via some PCIe bus bridge is further available, according to board block diagrams that can be found on the web.

All Tesla Motors vehicles manufactured from mid-October 2016 include a Drive PX 2, which will be used for neural net processing to enable Enhanced Autopilot and full self-driving functionality.[9] Other applications are Roborace.[10] Disassembling the Nvidia-based control unit from a recent Tesla car showed that a Tesla was using a modified single-chip Drive PX 2 AutoCruise, with a GP106 GPU added as a MXM Module. The chip markings gave strong hints for the Tegra X2 Parker as the CPU SoC.[11][12]

Volta based

Systems based on the Volta GPU microarchitecture were first announced at CES 2017[13]

Drive PX Xavier

The first Volta based Drive PX system was announced at CES 2017 as the Xavier AI Car Supercomputer.[13] It was re-presented at CES 2018 as Drive PX Xavier.[14][15] Initial reports of the Xavier SoC suggested a single chip with similar processing power to the Drive PX 2 Autochauffeur system.[16] However, in 2017 the performance of the Xavier-based system was later revised upward, to 50% greater than Drive PX 2 Autochauffeur system.[13] Drive PX Xavier is supposed to deliver 30 INT8 TOPS of performance while consuming only 30 watts of power.[17] This spreads across two distinct units, the iGPU with 20 INT8 TOPS as published early and the somewhat later on announced, newly introduced DLA that provided an additional 10 INT8 TOPS.

Drive PX Pegasus

In October 2017 Nvidia and partner development companies announced the Drive PX Pegasus system, based upon two Xavier CPU/GPU devices and two post-Volta (Turing) generation GPUs. The companies stated the third generation Drive PX system would be capable of Level 5 autonomous driving, with a total of 320 INT8 TOPS of AI computational power and a 500 Watts TDP.[18][19]

Ampere based

Drive AGX Orin

The Drive AGX Orin board family was announced on December 18, 2019 at GTC China 2019.[20] On May 14, 2020 Nvidia announced that Orin would be utilizing the new Ampere GPU microarchitecture and would begin sampling for manufacturers in 2021 and be available for production in 2022.[21] Follow up variants are expected to be further equipped with chip models and/or modules from the Tegra Orin SoC.

Comparison
Nvidia provided
reference board		 Drive CX Drive PX Drive PX 2

(AutoCruise)

 Drive PX 2

(Tesla)

 Drive PX 2

(AutoChauffeur)

 Drive PX 2

(Tesla 2.5)

 Drive PX Xavier[15] Drive PX Pegasus[22] Drive AGX Orin[20]

(compact / + GPU)

GPU Microarchitecture Maxwell (28 nm) Pascal (16 nm) Volta (12 nm) Ampere (7 nm)
Introduced January 2015 September 2016[23] October 2016[24] January 2016 August 2017[25] January 2017 October 2017 December 2019
Launched N/A N/A N/A N/A N/A N/A N/A N/A
Chips 1x Tegra X1 2x Tegra X1 1x Tegra X2 (Parker)

+ 1x Pascal GPU

 2x Tegra X2 (Parker)

+ 2x Pascal GPU

 2x Tegra X2 (Parker)

+ 1x Pascal GPU[26]

 1x Tegra Xavier[27] 2x Tegra Xavier

+ 2x Turing GPU

 2x Tegra Orin 2x Tegra Orin

+ 2x Next-Gen GPU

CPU 4x Cortex A57

4x Cortex A53

 8x Cortex A57

8x Cortex A53

 2x Denver

4x Cortex A57

 4x Denver

8x Cortex A57

 4x Denver

8x Cortex A57

 8x Carmel ARM64[27] 16x Carmel ARM64 12x Arm Hercules 24x Arm Hercules
GPU 2 SMM Maxwell

256 CUDA cores

 4 SMM Maxwell

512 CUDA cores

 1x Parker GPGPU

(1x 2 SM Pascal,
256 CUDA cores)

 1x Parker GPGPU

(1x 2 SM Pascal,
256 CUDA cores
on a MXM slot[11])

 2x Parker GPGPU

(2x 2 SM Pascal,
512 CUDA cores)
+ 2x dedicated MXM modules[28]

 1x Parker GPGPU

1x 2 SM Pascal,
256 CUDA cores [25][26]

 1x Volta iGPU

(512 CUDA cores)[27]

 2x Volta iGPU

(512 CUDA cores)
2x Turing dGPUs
(? CUDA cores)

 2x Ampere iGPU

(?CUDA cores)

 2x Ampere iGPU

(? CUDA cores)
2x Ampere dGPU
(? CUDA cores)

Accelerator 1x DLA

1x PVA[27]

 2x DLA

2x PVA

 2x DLA

2x PVA

 2x DLA

2x PVA

Memory 8GB LPDDR4[29] 16GB LPDDR4[29] LPDDR4[27] LPDDR5?
Storage 64GB eMMC[29] 128GB eMMC[29]
Performance 4 FP32 TFLOPS

10-12 DL TOPS[30][31]

 4 FP32 TFLOPS

10-12 DL TOPS [30][31]

 16 FP16 TFLOPS

8 FP32 TFLOPS

20-24 DL TOPS [30][31]

 4 FP32 TFLOPS

10-12 DL TOPS [30][31]

 20 INT8 TOPS, 1.3 FP32 TFLOPS (GPU)
10 INT8 TOPS, 5 FP16 TFLOPS (DLA)[27]		 320 INT8 TOPS (total)[32] 400 INT8 TOPS (total) 2000 INT8 TOPS (total)
TDP 20W[31] 40W

SoC portion: 10 W[23]

 40W

SoC portion: 10 W[23]

 80W[33][34][31][35]

SoC portion: 20 W[23]

 60W [33][34][31]

SoC portion: 20 W[23]

 30W[27] 500W[32] 130W 750W

Note: dGPU and memory are stand-alone semiconductors; all other components, especially ARM cores, iGPU and DLA are integrated components of the listed main computing device(s)

References
  1. Umar Zakir Abdul, Hamid; et al. (2016). "Current Collision Mitigation Technologies for Advanced Driver Assistance Systems–A Survey". PERINTIS eJournal. 6 (2). Retrieved 14 June 2017.
  2. NVIDIA DRIVE
  3. "Cars drive autonomously with Nvidia X1-based computer". Cnet. Cnet. 5 January 2015. Retrieved 29 March 2016.
  4. "Nvidia Announces Another Car 'Supercomputer' at CES". The Wall Street Journal. 4 January 2016. Retrieved 29 March 2016.
  5. Smith, Joshua Ho, Ryan. "NVIDIA Tegra X1 Preview & Architecture Analysis". Retrieved 2016-09-18.
  6. NVIDIA ebnet den Weg für die Autos von Morgen mit den NVIDIA-DRIVE-Automotive-Computern
  7. Smith, Ryan. "NVIDIA Announces DRIVE PX 2 – Pascal Power For Self-Driving Cars". Retrieved 2016-09-18.
  8. "Autonomous Car Development Platform from NVIDIA DRIVE PX2". www.nvidia.com. Retrieved 2016-09-18.
  9. Lambert, Fred (21 October 2016). "All new Teslas are equipped with NVIDIA's new Drive PX 2 AI platform for self-driving". Electrek. Retrieved 25 January 2017.
  10. Dow, Jameson (20 May 2017). "Roborace debuts their driverless "Robocar" on track at the Paris ePrix". Electrek. Retrieved 21 May 2017.
  11. Look inside Tesla’s onboard Nvidia supercomputer for self-driving
  12. Why Tesla’s Nvidia Supercomputer for Self-Driving is not as Powerful as Expected
  13. Cutress, Ian; Tallis, Billy (4 January 2016). "CES 2017: Nvidia Keynote Liveblog". Anandtech.com. Retrieved 9 January 2017.
  14. Baldwin, Roberto (8 January 2018). "NVIDIA unveils its powerful Xavier SOC for self-driving cars". Engadget. Retrieved 8 January 2018.
  15. Autonomous Car Development Platform from NVIDIA DRIVE PX2
  16. Smith, Ryan (28 September 2016). "Nvidia Teases Xavier, A High Performance SoC for Drive PX & AI". Anandtech. Retrieved 22 June 2017.
  17. "Autonomous Car Development Platform from NVIDIA DRIVE PX2". NVIDIA. Retrieved 2017-12-01.
  18. Oh, Nate. "NVIDIA Announces Drive PX Pegasus at GTC Europe 2017: Level 5 Self-Driving Hardware, Feat. Post-Volta GPUs". Retrieved 2017-10-10.
  19. Auchard, Eric (10 October 2017). "NVIDIA unveils next-generation platform for fully autonomous cars". Reuters. Retrieved 17 October 2017.
  20. Smith, Ryan. "NVIDIA Details DRIVE AGX Orin: A Herculean Arm Automotive SoC For 2022". www.anandtech.com. Retrieved 2019-12-21.
  21. Eric Abent (May 14, 2020). "NVIDIA Drive upshifts to Ampere GPUs for a smoother self-driving roadmap". Slash Gear.
  22. Oh, Nate. "NVIDIA Announces Drive PX Pegasus at GTC Europe 2017: Level 5 Self-Driving Hardware, Feat. Post-Volta GPUs". Retrieved 2017-10-10.
  23. Nvidia unveils palm-sized single SoC version of the DRIVE PX2
  24. "All Tesla Cars Being Produced Now Have Full Self-Driving Hardware". www.tesla.com. Retrieved 2017-12-01.
  25. "Tesla has a new Autopilot '2.5' hardware suite with more computing power for autonomous driving". Electrek. 2017-08-09. Retrieved 2018-10-26.
  26. "First look at Tesla's latest Autopilot (2.5) computer in Model 3, S, and X vehicles". Electrek. 2018-03-28. Retrieved 2018-10-26.
  27. NVIDIA Xavier SOC Is The Most Biggest and Complex SOC To Date
  28. NVIDIA Announces Pascal GPU Powered Drive PX 2 – 16nm FinFET Based, Liquid Cooled AI Supercomputer With 8 TFLOPs Performance
  29. "Hardware". NVIDIA Developer. 2017-03-30. Retrieved 2017-12-01.
  30. Ho, Joshua. "Hot Chips 2016: NVIDIA Discloses Tegra Parker Details". Retrieved 2018-10-26.
  31. Pirzada, Usman (2016-04-05). "Nvidia Drive PX 2 Uses Integrated and Discrete Pascal GPU Cores - 24 DL TOPS, 8 TFLOPs and Up To 4GB GDDR5 [Updated]". Wccftech. Retrieved 2018-10-26.
  32. "Nvidia AI and multi-petaOps chips for class 5 automated cars within 4 years - NextBigFuture.com". NextBigFuture.com. 2017-10-27. Retrieved 2018-10-26.
  33. Oberman, Stuart (October 2017). "NVIDIA GPU COMPUTING: A JOURNEY FROM PC GAMING TO DEEP LEARNING" (PDF).
  34. "thermal-design-power". devtalk.nvidia.com. Retrieved 2018-10-26.
  35. Altavilla, Dave. "NVIDIA Doubles Down On Self-Driving Cars With Xavier AI Chip And A Hat Tip To Next Gen Volta GPU". Forbes. Retrieved 2018-10-26.
This article is issued from Wikipedia. The text is licensed under Creative Commons - Attribution - Sharealike. Additional terms may apply for the media files.