MAROKO133 Breaking ai: ByteDance Introduces Astra: A Dual-Model Architecture for Autonomou

📌 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 Hot ai: Solid-state EV battery retains 78% capacity after 200 cycles u

Researchers at the Ulsan National Institute of Science and Technology (UNIST) have developed a method to extend the operational life of solid-state batteries by physically modifying the electrolyte layer. 

By applying uniaxial stretching to a fluorinated polymer electrolyte, the team achieved approximately 78% capacity retention after 200 charge-discharge cycles.

This performance marks a distinct improvement over the 55% retention rate observed in batteries utilizing unstretched electrolytes under identical conditions.

The study, published in the journal Energy Storage Materials, quantifies the impact of this technique on ion transport. 

“Experimental results showed that the lithium-ion diffusion rate in the stretched polymer electrolyte increased by 4.8 times compared to unstretched samples, with ionic conductivity improving by 72%,” said the researchers.

Enhanced safety features

Safety testing accompanied the performance evaluations. The research team confirmed that the new electrolyte possesses significant flame-retardant properties. 

During combustion tests, flames applied to the material extinguished within four seconds of ignition. 

This characteristic is relevant for the electric vehicle industry, where the flammability of current organic liquid electrolytes remains a primary safety concern.

Focus on structural arrangement

The technical approach focuses on the structural arrangement of the electrolyte material. The team used a fluorinated polymer film, PVDF-TrFE-CFE.

In its standard state, the polymer chains in this material are convoluted and entangled, hindering the efficient movement of lithium ions between the cathode and anode. The researchers applied a mechanical stretching process to align these polymer chains in a single direction.

“This physical stretching unfolds the convoluted polymer structure, opening up continuous pathways for lithium-ion movement,” explained the researchers.

“Additionally, incorporating ceramic powder (LLZTO) into the polymer matrix enhances mechanical flexibility, flame retardancy, and ion conductivity.”

The addition of this ceramic component is intended to enhance the mechanical flexibility of the film while also improving its flame-retardant properties and ionic conductivity.

Validation for commercialization

To validate the practical application of this material, the team integrated the stretched electrolyte into lithium-metal batteries equipped with lithium iron phosphate (LFP) cathodes. 

“When integrated into lithium-metal batteries with lithium iron phosphate (LFP) cathodes, the stretched electrolyte contributed to a notable increase in battery lifespan,” remarked the team.

“This research demonstrates that the inherent issues of polymer electrolytes—such as hindered lithium-ion transport—can be effectively addressed through a simple physical process like stretching,” explained Jonggeon Na, a researcher at the School of Energy and Chemical Engineering at UNIST and the study’s first author.

The study suggests that this processing technique could be adapted for mass production. Unlike inorganic solid electrolytes, which can be brittle and difficult to manufacture in large formats, polymer variants offer greater flexibility.

Professor Seok Ju Kang, who led the research, commented on the scalability of the findings. 

“Polymer electrolytes are more flexible and easier to produce at scale compared to inorganic solid electrolytes. The method developed in this study can be applied to various types of polymer electrolytes, accelerating the commercialization of safer, longer-lasting all-solid-state batteries,” Kang concluded.

🔗 Sumber: interestingengineering.com