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

📌 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

📌 MAROKO133 Breaking ai: Nerve-stimulating wearable targets heavy menstrual bleedi

Spark Biomedical’s neurostimulation wearable to reduce heavy menstrual bleeding debuts at CES 2026

Spark Biomedical is bringing its bioelectronic medicine research into menstrual health through OhmBody, its wellness brand and a wearable device of the same name, designed to reduce heavy menstrual bleeding using nerve stimulation rather than drugs or hormones.

Speaking at CES, Katherine Reil of OhmBody said the product builds directly on Spark Biomedicals’ earlier medical device work.

“Our company is Spark Biomedical.” She added, “Spark has an FDA-cleared device that helps with opioid withdrawal.”

That research laid the foundation for studying blood-loss control.

“We built off that research and started studying blood-loss reduction,” Reil said. “We originally thought about applying this to hemophilia.”

The company later shifted focus to menstrual health.

“Then we pivoted and asked, what if we apply this to periods something women experience every month?”

OhmBody is currently positioned as a wellness wearable.

“Ohm Body is a non-invasive, hormone-free wearable device,” Reil said. “It helps women experience more manageable, peaceful periods.”

Nerve signals control bleeding

The wearable works by stimulating specific nerves around the ear. “We stimulate the vagus and trigeminal nerves around the ear,” Reil said.

“That is called transcutaneous auricular neurostimulation.”

According to Spark Biomedical, this stimulation regulates the nervous system and influences blood flow.

“It regulates the nervous system,” Reil said. “It reduces blood loss by activating platelets in the spleen.”

The company recently published human trial data supporting this approach.

“We discovered blood-loss reduction by stimulating this nerve through our research,” Reil said. “We published those findings this year in Frontiers.”

The peer-reviewed pilot study included 16 participants, including women with von Willebrand Disease and those with heavy menstrual bleeding of unknown cause.

Participants using the OhmBody wearable daily during menstruation experienced an average reduction in menstrual blood loss of more than 50 percent and periods that were nearly 20 percent shorter.

Improvements were also reported in cramping, fatigue, and overall quality of life.

Spark Biomedical is continuing its clinical work. “We are running another clinical trial in February focused on heavy bleeding,” Reil said.

From wellness to medicine

While OhmBody is not yet FDA-cleared, the company says it is moving in that direction.

“It is a non-invasive, hormone-free option, which is really important,” Reil said. “I would say it is the first of its kind because it combines vagus and trigeminal nerve stimulation and impacts blood flow.”

Reil also pointed to longstanding gaps in women’s health research. “Women’s health has not been studied enough,” she said. “Talking about periods has been taboo.”

She added, “This is currently a wellness device, not FDA-cleared yet, but we are headed in that direction.”

Looking ahead, Spark Biomedical sees commercial expansion opportunities.

“In the next three to five years, we see a lot of opportunity through partnerships,” Reil said. “We want to expand into more e-commerce platforms and physical retail.”

The company is also targeting elite sports environments. “We would also like to enter the athlete space,” she said. “We would love to be in every Olympic training facility.”

Reil said CES provides a platform to push the conversation forward. “We are really thrilled to be here at CES,” she said. “We want to advocate for women through this work.”

You can explore all CES 2026 stories and coverage from the IE team by clicking here.

🔗 Sumber: interestingengineering.com