{"id":26290,"date":"2026-09-15T16:51:28","date_gmt":"2026-09-15T15:51:28","guid":{"rendered":"https:\/\/telecomkh.info\/?p=26290"},"modified":"2026-09-15T16:51:28","modified_gmt":"2026-09-15T15:51:28","slug":"physical-ai-takes-the-wheel-how-the-worlds-robotaxi-leaders-are-building-with-nvidia-technologies","status":"publish","type":"post","link":"https:\/\/telecomkh.info\/?p=26290","title":{"rendered":"Physical AI takes the wheel: How the world\u2019s robotaxi leaders are building with NVIDIA technologies"},"content":{"rendered":"

NVIDIA\u2019s modular, full-stack robotaxi pipeline \u2014 a three-computer solution spanning AI training, simulation and in-vehicle computing \u2014 is being adopted across the robotaxi ecosystem<\/strong><\/p>\n

 <\/p>\n

By Ali Kani, Nvidia<\/em><\/span><\/p>\n

The global robotaxi market \u2014 physical AI\u2019s first commercial breakthrough \u2014 is projected to reach $400 billion by 2035, with over 6 million commercial vehicles in operation as driverless fleets are already moving people through some of the world\u2019s busiest and most complex streets.
\nDeploying a driverless vehicle is one challenge. Scaling a fleet is a next-level computing challenge; it means delivering the same safe, reliable performance across thousands of vehicles.
\nMeeting those demands requires enormous amounts of compute across the robotaxi development lifecycle, from preparing and training AI models to simulating and validating driving behavior, as well as real-time processing in the vehicle.
\nNVIDIA provides an open platform for AI training, simulation and safety validation, with libraries, software development kits, workflows and models that developers can use alongside their own technology stacks.
\nEvery major robotaxi program operating at commercial scale today is running on NVIDIA\u2019s modular stack, spanning AI training, simulation, in-vehicle computing \u2014 or a combination of the three \u2014 to develop and deploy fleets at scale.<\/p>\n

What Is a Robotaxi Technology Stack?<\/strong>
\nA robotaxi technology stack is the end-to-end set of technologies used to develop, validate and deploy autonomous vehicles (AVs) \u2014 from data and AI model training to simulation, safety validation and real-time in-vehicle computing.
\nNVIDIA\u2019s robotaxi and AV platform brings these capabilities together in a three-computer solution: the model training computer, simulation and validation computer, and in-vehicle computer.<\/p>\n

1. Training Computer: NVIDIA DGX<\/strong>
\nRobotaxi intelligence advances as programs turn growing volumes of fleet data into increasingly capable models. Driving models can be trained on NVIDIA DGX systems.
\nThe NVIDIA Alpamayo portfolio of open reasoning vision language action (VLA) models, simulation frameworks and physical AI datasets gives developers building blocks they can adapt to their own data, requirements and technology stacks. Its reasoning models help address long-tail AV challenges by breaking complex driving situations into smaller steps, reasoning through each one and selecting the safest trajectory.
\nNVIDIA also provides physical AI datasets, reinforcement learning blueprints and recipes for post-training and distillation, helping developers optimize models for their target vehicles.
\nOn a challenging autonomous driving evaluation, adding meta-action and chain-of-thought reasoning data improved a VLA model\u2019s trajectory prediction accuracy, reducing minimum average displacement error \u2014 the predicted path\u2019s average deviation from the reference route \u2014 by 43%, from 2.08 to 1.18.<\/p>\n

2. Simulation and Validation Computer: NVIDIA Omniverse and Cosmos on NVIDIA RTX PRO<\/strong>
\nRobotaxi programs can\u2019t rely on physical miles alone to capture rare, long-tail driving scenarios. NVIDIA Omniverse NuRec models reconstruct real-world driving scenarios from sensor data, while NVIDIA Cosmos world foundation models generate physically based variations of them, enabling developers to turn thousands of real-world corner cases into millions of combinations of driving behavior, traffic, weather, lighting and sensor conditions.
\nFrom real-world corner cases to thousands of synthetic permutations spanning behavior and content, NVIDIA Cosmos variations expand AV training data and accelerate model deployment.
\nRunning on NVIDIA RTX PRO Servers, NVIDIA Omniverse and Cosmos support closed-loop simulation and validation. The NVIDIA AlpaSim simulation framework extends the workflow for training and evaluating reasoning-based autonomous-driving models, helping developers identify weaknesses before deployment.<\/p>\n

3. In-Vehicle Computer and Sensor Architecture: NVIDIA DRIVE Hyperion With DRIVE AGX<\/strong>
\nNVIDIA DRIVE Hyperion is NVIDIA\u2019s modular in-vehicle compute and sensor reference architecture for level-4-ready robotaxis. DRIVE Hyperion 10 pairs dual NVIDIA DRIVE AGX Thor systems-on-a-chip, built on the NVIDIA Blackwell platform, with 14 high-definition cameras, nine radars, three lidars and 12 ultrasonics for real-time, 360-degree sensor fusion. Its redundant compute and sensing design supports fail-operational driving if a sensor or compute component fails.
\nThe dual DRIVE AGX Thor is designed to run modern AI workloads \u2014 including VLA models \u2014 for perception, reasoning, path planning and driving actions.<\/p>\n

NVIDIA Halos provides a production-ready safety foundation through Halos OS, and a broader validation and certification framework spanning independent inspection, system validation, large-scale simulation and continuous testing from cloud to car.<\/p>\n

Robotaxi Leaders Adopting NVIDIA\u2019s Robotaxi Technology Stack<\/strong>
\nNVIDIA\u2019s robotaxi ecosystem spans every region where commercial robotaxi services are emerging today: Asia, Europe, the Middle East and North America. Across these markets, mobility providers, AV developers and automakers are adopting NVIDIA\u2019s three-computer architecture to train AI models, simulate and validate driving behavior, and deploy autonomous vehicles at scale.<\/p>\n

Scaling Robotaxi Services Globally<\/strong>
\n\u2022 Uber is scaling its fleet of NVIDIA DRIVE Hyperion, with plans to reach 28 cities by 2028. Uber and NVIDIA are also building a robotaxi AI data factory on NVIDIA Cosmos to curate fleet driving data for rare scenarios. Together, Uber and NVIDIA are collaborating with Autobrains, Avride, Lucid, May Mobility, Mercedes-Benz, Momenta, Nissan, Nuro, Pony.ai, Stellantis, Waabi, Wayve, WeRide and Zoox to bring NVIDIA-powered robotaxi services to the Uber platform.
\n\u2022 May Mobility is planning to operate autonomous ride-hailing services through Uber\u2019s network, while developing its software stack on the NVIDIA DRIVE platform.
\n\u2022 Bolt uses NVIDIA technologies to develop and scale AVs across Europe.
\n\u2022 Lyft plans to use NVIDIA DRIVE Hyperion as a reference architecture for future autonomous fleets. May Mobility vehicles are also currently operating on Lyft\u2019s network in Atlanta, powered by NVIDIA DRIVE.
\n\u2022 Through its partnership with Grab, WeRide plans to bring its DRIVE Hyperion- and DRIVE AGX Thor-based GXR to key markets across Southeast Asia.
\n\u2022 Waymo partners with NVIDIA to help build its autonomous computing system.<\/p>\n

Building Robotaxi Intelligence<\/strong>
\nBehind these services, AV developers are using NVIDIA accelerated computing, simulation and in-vehicle platforms to build the intelligence that operators and automakers deploy.
\n\u2022 Wayve, Nissan and Uber are developing a global robotaxi program using a prototype vehicle that combines Nissan\u2019s vehicle engineering, Wayve\u2019s embodied AI and the NVIDIA DRIVE Hyperion platform.
\n\u2022 Autobrains is developing robotaxi programs with Uber in Munich and VinFast in Southeast Asia, built on NVIDIA DRIVE Hyperion and enabled by Autobrains\u2019 Agentic AI technology.
\n\u2022 Zoox uses NVIDIA DRIVE for in-vehicle computing and cloud-based training and simulation.
\n\u2022 Momenta is developing its software stack based on NVIDIA DRIVE AGX running on DriveOS.
\n\u2022 Pony.ai developed its new-generation autonomous-driving domain controller with NVIDIA DRIVE Hyperion and DRIVE AGX Thor.
\n\u2022 Tensor is developing its level 4 Robocar with eight NVIDIA DRIVE AGX Thor systems-on-a-chip in its in-vehicle supercomputer.
\n\u2022 Waabi expands into the robotaxi market through a deployment collaboration with Uber; its Waabi Driver platform is built on NVIDIA DRIVE AGX Thor.
\n\u2022 TIER IV and Isuzu are deploying level 4 autonomous buses built on NVIDIA DRIVE Hyperion and DRIVE AGX Thor.
\n\u2022 Lenovo is supplying its NVIDIA DRIVE AGX Thor-based AD1 level 4 domain controller for a next-generation robotaxi program with SWM.
\n\u2022 DeepRoute.ai is developing a new generation of robotaxis built on the NVIDIA DRIVE Hyperion platform with DRIVE AGX Thor.<\/p>\n

Bringing Robotaxis Into Production<\/strong>
\nAs these systems move from development into production, automakers are integrating NVIDIA technology into autonomous and robotaxi-ready vehicle programs.
\n\u2022 Tesla trains its autonomous-driving neural networks on NVIDIA supercomputers.
\n\u2022 Mercedes-Benz and NVIDIA are collaborating with Uber to develop a robotaxi ecosystem based on the new S-Class, built on the NVIDIA DRIVE Hyperion architecture, full-stack NVIDIA DRIVE AV L4 software, and NVIDIA Alpamayo open AI models, simulation tools and datasets to support reasoning-based, safety-first autonomy.
\n\u2022 Stellantis, Wayve and Uber are collaborating to develop and deploy L4 driverless mobility services, leveraging NVIDIA DRIVE Hyperion and AI computing technologies.
\n\u2022 Lucid, Nuro and Uber are developing a global robotaxi service using NVIDIA DRIVE AGX Thor, part of the DRIVE Hyperion platform.
\n\u2022 Hyundai Motor and Kia are expanding their collaboration with NVIDIA to develop data-driven autonomous-driving systems built on NVIDIA DRIVE Hyperion. NVIDIA will also explore expanded collaboration with Hyundai Motor Group\u2019s joint venture, Motional, to advance level 4 robotaxi services.
\n\u2022 Geely, alongside its ecosystem partners, plans to develop and commercialize robotaxis using DRIVE Hyperion.
\n\u2022 Zeekr, a Geely Auto Group brand, has adopted DRIVE AGX Thor for a centralized domain controller.
\nFrom cloud to car, nearly every layer of the robotaxi platform is being developed on NVIDIA accelerated computing.<\/p>\n

Above, image credited to Nvidia<\/em><\/span><\/p>\n","protected":false},"excerpt":{"rendered":"

NVIDIA\u2019s modular, full-stack robotaxi pipeline \u2014 a three-computer solution spanning AI training, simulation and in-vehicle computing \u2014 is being adopted across the robotaxi ecosystem   By Ali Kani, Nvidia The global robotaxi market \u2014 physical AI\u2019s first commercial breakthrough \u2014 is projected to reach $400 billion by 2035, with over 6 million commercial vehicles in … <\/p>\n

Continue reading \u00abPhysical AI takes the wheel: How the world\u2019s robotaxi leaders are building with NVIDIA technologies\u00bb<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":26291,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[41,9,50],"tags":[],"_links":{"self":[{"href":"https:\/\/telecomkh.info\/index.php?rest_route=\/wp\/v2\/posts\/26290"}],"collection":[{"href":"https:\/\/telecomkh.info\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/telecomkh.info\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/telecomkh.info\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/telecomkh.info\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=26290"}],"version-history":[{"count":1,"href":"https:\/\/telecomkh.info\/index.php?rest_route=\/wp\/v2\/posts\/26290\/revisions"}],"predecessor-version":[{"id":26292,"href":"https:\/\/telecomkh.info\/index.php?rest_route=\/wp\/v2\/posts\/26290\/revisions\/26292"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/telecomkh.info\/index.php?rest_route=\/wp\/v2\/media\/26291"}],"wp:attachment":[{"href":"https:\/\/telecomkh.info\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=26290"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/telecomkh.info\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=26290"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/telecomkh.info\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=26290"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}