Category: Self-Driving & Autonomous Vehicles

Check out HALP (Hardware-Aware Latency Pruning), a new method designed to adapt convolutional neural networks (CNNs) and #transformer-based architectures for real-time performance. HALP optimizes pre-trained models to maximize compute utilization. In testing with NVIDIA DRIVE Orin™ on the road, it consistently outperformed alternative approaches. 00:00:00 – Introducing Hardware-Aware Latency Pruning (HALP) 00:00:29 – Common Model…

Humans naturally perceive the 3D geometry of objects and scenes in order to make decisions. Likewise, in the world of #autonomousdriving, the concept of “3D occupancy prediction” is critical to the development of safe and robust self-driving systems. In this episode of #DRIVELabs, we go beyond the traditional bird’s eye view approach and showcase NVIDIA’s…

Early Grid Fusion (EGF) is a new technique that enhances near-field obstacle avoidance in automatic parking assist. EGF combines machine-learned cameras and ultrasonic sensors to accurately detect and perceive surrounding obstacles, providing a 360-degree surround view. EGF solves the challenges of parking in tight spaces by detecting obstacles at a very close range as well…

Precise environmental perception is critical for #autonomousvehicle (AV) safety, especially when handling unseen conditions. In this episode of DRIVE Labs, we discuss a Vision Transformer model called SegFormer, which generates robust semantic segmentation while maintaining high efficiency. This video introduces the mechanism behind SegFormer that enables its robustness and efficiency. 00:00:00 – Robust Perception with…

Testing autonomous vehicles (AVs) in potential near-accident scenarios is critical for evaluating safety, but is difficult and unsafe to do in the real world. In this episode of DRIVE Labs, we discuss a new method from NVIDIA researchers called STRIVE (Stress-Test Drive), which automatically generates potential accident scenarios in simulation for AVs. The key to…

Understanding speed limit signs may seem like a straightforward task, but it can quickly become more complex in situations in which different restrictions apply to different lanes, or when driving in a new country. This episode of #DRIVELabs shows how #AI-based live perception can help AVs better understand the complexities of speed limit signs, using…

Diverse and #RedundantSensors, such as camera and radar, are necessary for robust AV perception. However, #radar sensors that leverage only traditional processing may not be fully up to the task. In this DRIVE Labs video, we show how AI can address the shortcomings of traditional radar signal processing in distinguishing both moving and stationary objects…

In this DRIVE Labs episode, we show how DRIVE IX perceives driver attention, activity, emotion, behavior, posture, speech, gesture and mood. Driver perception is a key aspect of the platform that enables the AV system to ensure a driver is alert and paying attention to the road. It also enables the AI system to perform…

In this DRIVE Labs episode, we show how software-defined AI techniques can be used to significantly improve performance and functionality of our light source perception deep neural network (DNN) — increasing range, adding classification capabilities and more — in a matter of weeks. Learn more: https://news.developer.nvidia.com/optimizing-light-source-perception-with-software-defined-ai/ Watch the full series here: https://nvda.ws/2Ru1yDL Learn more about…

High-definition (HD) maps are detailed digital models of the world, including elements like road geometry, signs, and traffic signals. While they’ve been widely used for autonomous driving, they’re costly to create and maintain. This has led to the growing demand for “mapless driving.” In this episode of DRIVE Labs, we’ll explore NVIDIA innovations that enhance…