MAROKO133 Eksklusif ai: ByteDance Introduces Astra: A Dual-Model Architecture for Autonomo

📌 MAROKO133 Eksklusif ai: ByteDance Introduces Astra: A Dual-Model Architecture fo

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

📌 MAROKO133 Eksklusif ai: Trump’s Huge AI Project Is Running Into a Major Financia

It’s 2026, and tech companies continue to insist they need to spend staggering amounts of money on AI data centers. Yet for all of its enthusiasm over the past few years, Wall Street is finally starting to squint at the numbers.

According to new reporting by Business Insider, JPMorgan Chase is running into trouble finding investors interested in servicing billions in debt backing two of the first five Stargate data centers.

Stargate is Donald Trump’s $500 billion AI project led by tech companies Oracle and OpenAI. Its vague goal, OpenAI has explained, is to “secure American leadership in AI,” which will somehow “support the re-industrialization of the United States” and “provide a strategic capability to protect the national security of America and its allies.”

But if JPMorgan — which led financial lenders through a $38 billion Stargate debt raise — is already struggling to sell the vision, that all might be easier said than done. Per BI, a person familiar with both data centers said they’re fully financed, though noted that banks and other investors are growing jittery about pouring even more money into the megaproject.

“We are hearing from market participants that in some cases, there may be banks that could be approaching the exposure levels they’re comfortable with when it comes to certain data center projects,” S&P Global Infrastructure Ratings director Dhaval Shah told BI.

That could mean trouble for Oracle, the cloud gianttapped to provide the physical infrastructure for Stargate. Back in September, S&P Global noted it was considering cutting Oracle’s credit rating — a letter-grade investors use to judge a company’s creditworthiness — below BBB.

The scale runs from AAA at the very top to D, or default, at the bottom. Anything below BBB- would place Oracle in junk-bond territory, which would significantly raise the cost to borrow more debt. In other words, Stargate may soon be seen as so risky that the financial situation behind its data centers becomes unsustainable.

To put the trouble with the two initial data centers into perspective, Trump said last year that Stargate could require as many as 20 data centers when all is said and done. If the market is already reaching its comfortable limit, it’s difficult to imagine a world where Stargate proceeds without significant revisions.

“I am very surprised these loans were even underwritten at the time,” Gil Luria, an analyst at the firm DA Davidson told BI, referring to the data centers that have been funded. “The market has indicated this is not investment-grade debt.”

More on data centers: AI Data Centers Are an Even Bigger Disaster Than Previously Thought

The post Trump’s Huge AI Project Is Running Into a Major Financial Problem appeared first on Futurism.

🔗 Sumber: futurism.com