MAROKO133 Breaking ai: Uber-backed robotaxi to hit public roads as Lucid, Nuro advance aut

📌 MAROKO133 Breaking ai: Uber-backed robotaxi to hit public roads as Lucid, Nuro a

Lucid Group, Nuro, and Uber have unveiled the production-intent robotaxi that will power their upcoming global autonomous ride-hailing service. The vehicle and its Uber-designed in-cabin rider experience were revealed at CES 2026 in Las Vegas.

The companies said the robotaxi is designed to deliver a premium passenger experience while supporting large-scale autonomous deployment. It is built on the all-electric Lucid Gravity platform and integrates Nuro Level 4 autonomous driving technology with Uber ride-hailing operations.

The companies also confirmed that autonomous on-road testing began in December. The testing milestone marks a key step toward launching the robotaxi service in the San Francisco Bay Area later in 2026.

Nuro is leading the testing using robotaxi engineering prototypes supervised by autonomous vehicle operators. The goal is to validate performance, safety, and reliability in real-world driving conditions ahead of commercial launch.

Sensors define the ride

The robotaxi features a next-generation sensor array that provides 360-degree perception. It combines high-resolution cameras, solid-state lidar sensors, and radars integrated across the vehicle body and a purpose-built roof-mounted halo.

The low-profile halo is designed to preserve the Lucid Gravity design while supporting advanced sensing. Integrated LEDs help riders identify the correct vehicle, display rider initials, and communicate ride status from pickup through drop-off.

Inside the cabin, the robotaxi offers an interactive rider experience focused on comfort and transparency. Screens allow passengers to control heated seats, climate settings, and music, contact support, or request the vehicle to pull over.

Riders can also view real-time visualizations showing what the robotaxi sees and how it plans its path. This includes actions such as yielding to pedestrians, stopping at traffic lights, changing lanes, and completing drop-offs.

High-performance computing is powered by NVIDIA DRIVE AGX Thor, part of the NVIDIA DRIVE Hyperion platform. The system supports real-time AI processing and integration required for Level 4 autonomous driving.

“The debut of our production intent robotaxi with Lucid and Uber is a significant milestone on our path to delivering autonomy at scale,” said Dave Ferguson, Co-Founder and Co-CEO of Nuro. “By bringing together Nuro’s proven level 4 autonomy, Lucid’s advanced vehicle architecture, and Uber’s global reach, we’re building a robotaxi service designed for real-world operations and long-term growth.”

Testing before scaling up

Nuro said autonomous on-road testing is part of its safety and validation framework developed through years of commercial autonomous deployments. The program evaluates dozens of capabilities across the full autonomy stack.

Testing includes closed-course validation and large-scale simulation alongside public road testing. The process is designed to assess performance across a wide range of traffic and environmental scenarios.

“Uber is proud to partner with Lucid and Nuro to bring a state-of-the-art robotaxi to market later this year,” said Sarfraz Maredia, Global Head of Autonomous Mobility & Delivery at Uber.

“By combining leading expertise in electric vehicles, autonomy, and ride-hailing, we’re building a unique new option for affordable and scalable autonomous rides in the San Francisco Bay Area and beyond.”

“Our robotaxi program with Uber and Nuro is a key part of how Lucid is leveraging its technology to create a more sustainable future of mobility that is widely accessible,” said Kay Stepper, Vice President of ADAS and Autonomous at Lucid.

Pending final validation, the production-intent robotaxi is expected to enter production at Lucid’s Arizona factory later this year. The vehicle is on public display at the NVIDIA CES showcase through January 8.

For ongoing news, in-depth reporting, and key developments from CES 2026, read the IE team’s coverage here. 

🔗 Sumber: interestingengineering.com

📌 MAROKO133 Update ai: ByteDance Introduces Astra: A Dual-Model Architecture for A

The increasing integration of robots across various sectors, from industrial manufacturing to daily life, highlights a growing need for advanced navigation systems. However, contemporary robot navigation systems face significant challenges in diverse and complex indoor environments, exposing the limitations of traditional approaches. Addressing the fundamental questions of “Where am I?”, “Where am I going?”, and “How do I get there?”, ByteDance has developed Astra, an innovative dual-model architecture designed to overcome these traditional navigation bottlenecks and enable general-purpose mobile robots.

Traditional navigation systems typically consist of multiple, smaller, and often rule-based modules to handle the core challenges of target localization, self-localization, and path planning. Target localization involves understanding natural language or image cues to pinpoint a destination on a map. Self-localization requires a robot to determine its precise position within a map, especially challenging in repetitive environments like warehouses where traditional methods often rely on artificial landmarks (e.g., QR codes). Path planning further divides into global planning for rough route generation and local planning for real-time obstacle avoidance and reaching intermediate waypoints.

While foundation models have shown promise in integrating smaller models to tackle broader tasks, the optimal number of models and their effective integration for comprehensive navigation remained an open question.

ByteDance’s Astra, detailed in their paper “Astra: Toward General-Purpose Mobile Robots via Hierarchical Multimodal Learning” (website: https://astra-mobility.github.io/), addresses these limitations. Following the System 1/System 2 paradigm, Astra features two primary sub-models: Astra-Global and Astra-Local. Astra-Global handles low-frequency tasks like target and self-localization, while Astra-Local manages high-frequency tasks such as local path planning and odometry estimation. This architecture promises to revolutionize how robots navigate complex indoor spaces.

Astra-Global: The Intelligent Brain for Global Localization

Astra-Global serves as the intelligent core of the Astra architecture, responsible for critical low-frequency tasks: self-localization and target localization. It functions as a Multimodal Large Language Model (MLLM), adept at processing both visual and linguistic inputs to achieve precise global positioning within a map. Its strength lies in utilizing a hybrid topological-semantic graph as contextual input, allowing the model to accurately locate positions based on query images or text prompts.

The construction of this robust localization system begins with offline mapping. The research team developed an offline method to build a hybrid topological-semantic graph G=(V,E,L):

• V (Nodes): Keyframes, obtained by temporal downsampling of input video and SfM-estimated 6-Degrees-of-Freedom (DoF) camera poses, act as nodes encoding camera poses and landmark references.
• E (Edges): Undirected edges establish connectivity based on relative node poses, crucial for global path planning.
• L (Landmarks): Semantic landmark information is extracted by Astra-Global from visual data at each node, enriching the map’s semantic understanding. These landmarks store semantic attributes and are connected to multiple nodes via co-visibility relationships.

In practical localization, Astra-Global’s self-localization and target localization capabilities leverage a coarse-to-fine two-stage process for visual-language localization. The coarse stage analyzes input images and localization prompts, detects landmarks, establishes correspondence with a pre-built landmark map, and filters candidates based on visual consistency. The fine stage then uses the query image and coarse output to sample reference map nodes from the offline map, comparing their visual and positional information to directly output the predicted pose.

For language-based target localization, the model interprets natural language instructions, identifies relevant landmarks using their functional descriptions within the map, and then leverages landmark-to-node association mechanisms to locate relevant nodes, retrieving target images and 6-DoF poses.

To empower Astra-Global with robust localization abilities, the team employed a meticulous training methodology. Using Qwen2.5-VL as the backbone, they combined Supervised Fine-Tuning (SFT) with Group Relative Policy Optimization (GRPO). SFT involved diverse datasets for various tasks, including coarse and fine localization, co-visibility detection, and motion trend estimation. In the GRPO phase, a rule-based reward function (including format, landmark extraction, map matching, and extra landmark rewards) was used to train for visual-language localization. Experiments showed GRPO significantly improved Astra-Global’s zero-shot generalization, achieving 99.9% localization accuracy in unseen home environments, surpassing SFT-only methods.

Astra-Local: The Intelligent Assistant for Local Planning

Astra-Local acts as the intelligent assistant for Astra’s high-frequency tasks, a multi-task network capable of efficiently generating local paths and accurately estimating odometry from sensor data. Its architecture comprises three core components: a 4D spatio-temporal encoder, a planning head, and an odometry head.

The 4D spatio-temporal encoder replaces traditional mobile stack perception and prediction modules. It begins with a 3D spatial encoder that processes N omnidirectional images through a Vision Transformer (ViT) and Lift-Splat-Shoot to convert 2D image features into 3D voxel features. This 3D encoder is trained using self-supervised learning via 3D volumetric differentiable neural rendering. The 4D spatio-temporal encoder then builds upon the 3D encoder, taking past voxel features and future timestamps as input to predict future voxel features through ResNet and DiT modules, providing current and future environmental representations for planning and odometry.

The planning head, based on pre-trained 4D features, robot speed, and task information, generates executable trajectories using Transformer-based flow matching. To prevent collisions, the planning head incorporates a masked ESDF loss (Euclidean Signed Distance Field). This loss calculates the ESDF of a 3D occupancy map and applies a 2D ground truth trajectory mask, significantly reducing collision rates. Experiments demonstrate its superior performance in collision rate and overall score on out-of-distribution (OOD) datasets compared to other methods.

The odometry head predicts the robot’s relative pose using current and past 4D features and additional sensor data (e.g., IMU, wheel data). It trains a Transformer model to fuse information from different sensors. Each sensor modality is processed by a specific tokenizer, combined with modality embeddings and temporal positional embeddi…

Konten dipersingkat otomatis.

🔗 Sumber: syncedreview.com