📌 TOPINDIATOURS Update ai: Meta breaks ground on $10 billion, 1GW data center in I

Meta is breaking ground on a 1 gigawatt data center campus in Lebanon, Indiana, marking one of the company’s largest infrastructure investments to date.

The project represents more than $10 billion in spending on data center infrastructure and related community development.

The Lebanon site will be Meta’s second data center location in Indiana. Once fully operational, the campus is designed to deliver 1GW of capacity, giving the company flexibility to run both its core digital platforms and fast-growing AI workloads.

As AI systems demand more compute power, gigawatt-scale facilities are becoming critical.

Large campuses allow companies to centralize high-density computing while supporting higher bandwidth, lower latency and improved reliability.

Meta says the Indiana site is built with long-term expansion and technology shifts in mind. The scale is intended to support both existing services and future AI ambitions without building separate facilities for each.

Built for gigawatt scale

At peak construction, the project is expected to support more than 4,000 jobs. Once operational, about 300 permanent positions will remain at the site.

Beyond direct employment, Meta is launching a Boone County-wide workforce development initiative through the Boone County Career Collaborative.

The program will focus on career exploration and work-based learning for students across three school districts, connecting classrooms with local employers.

The company is also committing financial support to local residents. It will provide $1 million each year for 20 years to the Boone REMC Community Fund to help families with energy bills.

Additional funding will support emergency water utility assistance through The Caring Center.

Meta says it will pay the full cost of the energy, water and wastewater services required to run the facility.

Over the course of the project, the company plans to invest more than $120 million in critical water infrastructure in Lebanon, along with upgrades to roads, transmission lines and other utilities.

Powering AI responsibly

The data center will match 100 percent of its electricity use with clean energy, according to the company. It is also targeting LEED Gold certification once operational.

To reduce water use, the site will rely on a closed-loop, liquid-cooled system that recirculates the same water. Meta says the system will use zero water for a majority of the year.

The company also plans to restore 100 percent of the water it consumes at the Lebanon facility back to local watersheds.

As part of that effort, Meta is partnering with Arable to deploy irrigation technology for independent farmers in Indiana’s Upper Wabash River Basin.

The initiative is expected to restore 200 million gallons of water per year for ten years while lowering irrigation costs.

In addition, a section of Deer Creek will be revitalized to improve wetland health, expand vegetation and create better habitats for pollinators.

Meta says it aims to be a long-term partner in the region as the campus comes online, positioning the Lebanon site as both a compute hub for AI and a major economic engine for Boone County.

🔗 Sumber: interestingengineering.com

📌 TOPINDIATOURS Breaking ai: ByteDance Introduces Astra: A Dual-Model Architecture

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