Explore featured sessions for autonomous vehicles.

Discover proven techniques for developing safer, more efficient transportation. Attend talks and panels covering advancements in autonomous driving, end-to-end vehicle simulation, robotaxis, trucking, and more. In addition, NVIDIA DRIVE Developer Day offers several deep dive sessions on safe and robust autonomous vehicle development led by in-house experts.

Featured Speakers

Off-road and On-road: Building Industrial Autonomous Solutions and New Business Models

Nils Jaeger 
President and Head of Volvo Autonomous Solutions, Volvo Group

Accelerating the Path to Zero Collisions

Ödgärd Andersson
CEO, Zenseact

Developing Software-Defined Trucks for an Autonomous Future

Axel Gern
CTO Daimler Autonomous Technology Group, Daimler Trucks

Hard Miles Without Hard Miles

Paul Newman
Founder and CTO, Oxbotica

Sessions By Topic

Autonomous Vehicles (AV)

  • Fireside Chat: Software Platforms for Future Generations of Autonomous Vehicles

    • Guillaume Binet, Vice President of Software Infrastructure, Motional

    Guillame Binet, Vice President of Software Infrastructure at Motional, will discuss the robotaxi company's journey to fully driverless deployment, highlighting the company’s development of AI infrastructure used to build self-driving software that is truly safe and reliable.

  • The Road to Mass Deployment

    • Anurag Ganguli, VP of R&D, Plus

    It’s widely acknowledged that autonomous trucks will bring in transformational impacts to the transportation and logistics industry. While driverless trucks may be a few years away, highly automated driver-in systems can provide substantial benefits today. In this talk, we'll discuss the value propositions of such a driver-in product, which include fuel savings leading to reduced carbon emissions, increased driver comfort leading to improved driver retention, and overall increased safety. In addition, we’ll discuss the technological challenges associated with building an autonomous product that can be deployed at scale.

  • Fireside Chat: Harnessing Autonomous Technology to Shape Community Transportation

    • Sean Harrington, CEO, Optimus Ride
    • Gabe Klein, Founding Partner, CityFi

    Modern communities face transportation challenges, including critical mobility gaps, weak connection between people and communities, growing emissions, and traffic and parking congestion. Autonomous vehicles (AV) are a promising technology that, if applied correctly, can offer both sustainable, community-forward transportation solutions and immediate adoption and application of AV technology.This session will discuss how best to harness the power of autonomous technology to positively shape community transportation.

  • Making Sense of Behavioral Scenarios for Validation of Autonomous Vehicle Safety

    • Jim Cherian, Senior Research Fellow and Lead (AV Virtual Validation), CETRAN (Center for Testing and Research on Autonomous Vehicles - NTU), Nanyang Technological University, Singapore

    Scenario-based testing (SbT) is commonly used to ascertain the behavioral safety of autonomous vehicles (AV) before public trials. However, quality of SbT test cases, objectivity of the evaluation metrics, and acceptance criteria of AV behavior are major challenges in achieving an effective SbT.

AV Simulation

  • DRIVE Sim on Omniverse

    • Matthew Cragun, Sr. Product Manager, NVIDIA

    NVIDIA built DRIVE Sim on the Omniverse platform, enabling new functionalities and new use cases. This presentation will showcase new software capabilities and discuss synthetic data generation for perception development, planning and control algorithm development, software integration, and end-to-end testing.

  • Architecting a New Platform for AV Simulation from the Ground Up

    • Alex Hyder, Director Product Development, DRIVE Sim, NVIDIA
    • Matt Campbell, Senior Engineering Manager, DRIVE Sim, NVIDIA

    Autonomous vehicle sensor simulation requires high levels of computation in order to generate realistic sensor data of virtual worlds. This is even more challenging when considering Hardware-in-the-Loop (HIL) tests where all of the data must be generated at real-time rates and in a repeatable way. We’ll discuss our experience building DRIVE Sim from the ground up on the Omniverse platform and lessons learned along the way.

  • ADAS and AV Safety - Dealing with an Infinite Test Space

    • Roy Fridman, VP Business Development, Foretellix

    The development of ever-improving smart and autonomous cars is a mammoth transformation challenge for OEMs, Tier 1s and the automotive ecosystem. The nature of ADAS and AV systems requires a paradigm shift in the way they’re verified and validated. Given the unprecedented level of complexity, traditional automotive testing approaches are no longer sufficient. This presentation will discuss some of the challenges of V&V and propose new ways to tackle them.

AV Trends

  • Automotive Software and Hardware Architecture - Innovation and Disruption Beyond the Car´s Chassis

    • Luca De Ambroggi, Chief Analyst, WARDS Auto - Informa

    Vehicles belong to the broad IoT world already, but in the future they’ll need to seamlessly and efficiently mesh, and share data, with a much more complex ecosystem. This includes domains like manufacturers´ back-end, public-cloud, connectivity infrastructure, and the fast-expanding “smart world”. In this session, we’ll dive into those topics, as well other trends in E/E architectures, software platforms, and centralized hardware.

  • From Automated Driving to ADAS - Leveraging The Technology for the Masses

    • Sam Abuelsamid, Principal Analyst, Guidehouse Insights

    As the timeline for volume deployment of highly automated driving (SAE L4/L5) has stretched further into the future, there’s still a strong near-term opportunity to get returns on the massive investments being made on AVs. The technologies that have evolved for sensing, compute platforms, and software for automated driving are increasingly being applied to improve capabilities of advanced driver assistance systems (ADAS). High-resolution cameras, thermal imaging, imaging radar, and lidar are all being incorporated into updated ADAS systems to enable proactive safety and improved convenience while high-performance SoCs are enabling centralized compute that includes machine learning.

  • Cloud Strategies to Accelerate Autonomous Vehicle Development and Validation

    • Vijitha Chekuri, Autonomous Business Development Leader, Amazon Web Services (AWS)
    • Norm Marks, Global Lead Automotive Enterprise

    Autonomous driving development presents a massive computational challenge. Petabytes of sensor data must be processed for training and testing, which impacts time-to-market, scale, and cost throughout the development cycle. Validating these systems requires large-scale testing in a wide range of rare and dangerous scenarios before deployment. In this session, Amazon Web Services and NVIDIA will explain how to reliably scale AV development and validation by using the latest technologies, including NVIDIA DriveWorks, AWS SageMaker, and NVIDIA DRIVE Sim, all available on AWS.

  • NVIDIA AGX Platform: Scalable and Modular Architecture for High-Performance AI Compute

    • Amit Badlani, Senior Product Manager, NVIDIA

    This session will provide an in-depth view into the NVIDIA AGX product line up based on the NVIDIA Orin SoC. It will feature the modular and scalable architecture of these products and dive deeper into how these are used across multiple industries such as autonomous vehicles, robotics, and healthcare.

AV Perception

  • Using Next-Generation Lidar on NVIDIA DRIVE

    • Amir Day, Director of Computer Vision, Innoviz Technologies

    Amir Day, Director of Computer Vision at Innoviz Technologies, will discuss the next generation of lidar sensors and introduce the new InnovizTwo LiDAR, which offers an extensive set of features and high performance at a reduced cost. We'll also cover how Innoviz gathers data, develops deep learning capabilities, and is compatible with the high-performance, energy-efficient NVIDIA DRIVE platform.

  • Overcoming AV Challenges with AI-Based 4D Imaging Radar

    • Ian Podkamien, VP and Head of Automotive, Vayyar Imaging

    Limitations of current ADAS technologies are a significant hurdle to achieving truly driverless vehicles. In terms of software, the issues include system incompatibility, integration hurdles, and sheer code volume. This session will explain how ML-based 4D imaging radar addresses these obstacles and dive into how it simultaneously enables dozens of leading-edge SRR, MRR, and LRR applications, cutting complexity and reducing costs across the board.

  • Obstacle Detection Using Stereoscopic Vision Systems on NVIDIA DRIVE

    • David Lempert, VP R&D, Foresight Automotive

    Autonomous machines face two major challenges: detecting any obstacle on the road and safely operating in even extreme weather and lighting conditions. A vision solution that offers forward-facing detection of objects on the road can enable machine autonomy across multiple industries and applications. Stereoscopic vision technology uses two synchronized cameras to generate a depth map that enables machines to detect any object as well as its accurate size, location, and distance. We'll describe how NVIDIA's DRIVE platform is being used in Foresight's stereoscopic obstacle detection software, composed of both visible-light and infrared channels and designed to provide accurate information about any object in harsh weather and lighting conditions.

  • Using Raw Fusion for ADAS/AV Perception

    • Ronny Cohen, Senior Director Israel Operations and Perception, LeddarTech

    Perception for autonomous driving and advanced driver assistance systems (ADAS) must be reliable, efficient, and scalable. Achieving these requirements calls for a redundant and diverse sensor set. We'll cover how raw sensor fusion can support such a system for robust perception.

Special Event: DRIVE Developer Day

  • Leveraging DRIVE Hyperion for AV Development

    • Mark Fertelmeister, Manager of Product Management, NVIDIA

    Aligning to NVIDIA’s in-house DRIVE Hyperion 8.1 AV development solution provides a tremendous benefit to customers by enabling them to leverage an existing robust system architecture and sensor set. In this session, learn how to jumpstart AV development and save time and money with the DRIVE Hyperion platform.

  • NVIDIA DRIVE SDK: The Foundation of AV Development

    • Varun Murthy, Solutions Architect, NVIDIA

    NVIDIA DRIVE SDK is the cornerstone of AV development, creating a solid foundation for implementing autonomous vehicle software. The latest version includes NVIDIA DRIVE OS 6 and DriveWorks 5 SDKs, which unlock the accelerated hardware and high performance of NVIDIA DRIVE AGX Orin. This technical overview will give attendees a first-hand view into developing optimized AV applications with the latest DRIVE platform.

  • Optimizing Deep Neural Networks for NVIDIA DRIVE

    • Fabian Weise, Solutions Architect, NVIDIA

    Deep learning and deep neural networks (DNNs) are an integral part of state-of-the-art self-driving perception stacks. This session introduces running custom DNNs on NVIDIA DRIVE AGX, from searching and benchmarking networks to deploying them efficiently in the vehicle. We’ll also share suitable optimization strategies with a special focus on TensorRT—NVIDIA’s inference engine. Finally, an overview of the novel optimization features and methodology coming with the DRIVE Orin system-on-a-chip will be provided, including sparsity, tiling, and chaining.

  • Scaling ML Ops for AV Development

    • Clement Farabet, VP of AI Infrastructure, NVIDIA
    • Nicolas Koumchatzky, Senior Director of AI Infrastructure

    A primary challenge in developing autonomous vehicles is the complexity of the infrastructure and automation required to support AI development. The AI powering AVs needs to learn to predict hundreds of properties in real-world coordinates, which requires advanced machinery to feed it the right datasets. In this session, we focus on how NVIDIA is approaching MLOps at the scale necessary to deploy AVs, including the automation and workflows we’ve put together, from the fleet to the cloud, through our simulator, and back to the fleet.

  • Developing Automated Parking Technologies

    • Atousa Torabi, Product Manager for Isaac Robotics and Parking, NVIDIA

    Automatic parking encompasses fully automated parking in and out; capable of parallel, perpendicular, and angled parking. This feature must safely operate in a variety of parking conditions while in the presence of multiple road users and obstacles. This session will present the NVIDIA AI-based parking stack and how to use simulation and synthetic data for automatic parking development.

  • High-Definition, Scalable Mapping for Autonomous Vehicles

    • Rambo Jacoby, Principal Product Manager for Autonomous Driving, NVIDIA

    HD Maps are a central part of the AV experience. This session will dive into NVIDIA's strategy for delivering a comprehensive mapping solution to our AV partners, demonstrate how maps are used as part of our end-to-end AV solution, and discuss how our work is accelerated by the recent acquisition of DeepMap.

  • Automating AV Verification and Validation

    • Justyna Zander, Global Head of Verification and Validation for Autonomous Driving, NVIDIA
    • Ahmed Nassar, Head AV Verification and Validation Architect, NVIDIA

    This session will discuss the building blocks to achieving fully automated verification and validation of autonomous vehicles. We’ll briefly touch upon HSDL Scenario Generation and thoroughly describe the HSDL Observer Engine used both in simulation and replay. The ultimate objective is to provide a framework that allows developers to build self-validating AV driving functions.

  • Developing and Testing Autonomous Vehicles at Scale

    • Norm Marks, Senior Director of Automotive, NVIDIA
    • Rambo Jacoby, Principal Product Manager for Autonomous Driving

    Developing, training, and testing AI for autonomous driving is an extremely complex and costly effort. In this session, we ‘ll cover how this enterprise-ready platform speeds the development of complex AI models using DGX SuperPOD, NVIDIA Base Command, NVIDIA Fleet Command, and pretrained models from NVIDIA NGC.

See automotive/AV session highlights from the previous GTC. Get ready for what’s to come.

Find the complete automotive/AV GTC On-Demand playlist here.

Explore More Conference Topics

Explore All Session Topics

NVIDIA Developer Program

Get the advanced tools and training you need to successfully build applications on all NVIDIA technology platforms.

Accelerate your Startup

Explore the startup track at GTC to learn how NVIDIA Inception can fuel your growth through go-to-market support, world-class training, and technology assistance.

Get Hands-On Training

Interested in developing key skills in AI, accelerated data science, or accelerated computing? Get hands-on instructor-led training from the NVIDIA Deep Learning Institute (DLI) and earn a certificate demonstrating subject matter competency.