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

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

📌 TOPINDIATOURS Update ai: Sebestyen’s theorem crosses into infinity after 40 year

A decades-old rule in mathematics has just been pushed beyond its long-standing limits, opening new ground for how scientists describe the physical world.

At the University of Vaasa in Finland, mathematician Yosra Barkaoui has successfully generalized a fundamental theorem that had remained confined to “bounded” systems for more than 40 years.

Her doctoral dissertation extends Sebestyén’s theorem into the unbounded realm, a shift with deep implications for theoretical physics and advanced mathematics.

Unbounded operators are central to physics. They are used to describe quantities such as kinetic energy, momentum, and time: values that can grow without limit.

Yet, until now, key mathematical rules governing these operators lacked a rigorous foundation beyond bounded cases.

Barkaoui’s work focuses on nonnegative closed operators, mathematical objects that model real-world quantities that cannot drop below zero.

By extending Sebestyén’s theorem, first introduced in 1983, she has provided mathematicians with a more complete framework for understanding how these operators behave.

“The Sebestyén theorem has been around since 1983, but it was only explored in the bounded case,” said Barkaoui.

“This is the first time the theorem has been extended to the unbounded case and to linear relations.”

Beyond bounded mathematics

In mathematics, bounded operators have a finite “size,” or norm. These systems are easier to control and analyze, which is why much of the theory has focused on them.

Unbounded operators, by contrast, can grow infinitely large, making them far more complex and harder to handle.

Barkaoui’s research shows that rules developed for bounded systems do not automatically apply to unbounded ones.

Her work reveals that assumptions long taken for granted were being carried over incorrectly.

“Many models in physics are based on unbounded systems,” Barkaoui explained. “What is new in our work is that we found a link between two types of inequalities that describe how the underlying operators relate to each other.”

This newly identified connection helps explain how unbounded operators behave under different mathematical constraints, offering clarity in an area that has long challenged researchers.

Although the work is theoretical, it strengthens the foundations upon which applied mathematics and physics are built, making future discoveries more reliable.

Building stronger foundations

Barkaoui emphasizes that her findings are not about immediate applications, but about enabling deeper exploration.

“Our results give mathematicians the tools to work more confidently with unbounded operators,” she said. “When the theoretical foundation is clear, it becomes easier to explore new questions and make further discoveries.”

The dissertation also marks a personal milestone. It is Barkaoui’s second PhD in mathematics, following her first doctorate completed in Tunisia.

She chose to pursue another doctoral degree largely to work under the supervision of Professor Seppo Hassi at the University of Vaasa.

“It was a dream of mine to work with Professor Hassi,” Barkaoui said. “I truly admire him, both as a mathematician and as a person.”

She described the experience as both intellectually and personally rewarding, highlighting the role mentorship plays in advanced research.

“Working with him has been a real pleasure and a privilege, and his guidance has meant a great deal to me,” she said.

By extending a long-standing theorem into uncharted territory, Barkaoui’s work strengthens the mathematical backbone behind modern physics and abstract theory alike, reinforcing the importance of rigorous foundations in scientific progress.

🔗 Sumber: interestingengineering.com