MAROKO133 Update ai: ByteDance Introduces Astra: A Dual-Model Architecture for Autonomous

📌 MAROKO133 Hot ai: ByteDance Introduces Astra: A Dual-Model Architecture for Auto

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 Update ai: The Supreme Court Just Dealt a Crushing Blow to “AI Artists

Proponents of generative AI say the tech has greatly lowered the barriers of entry in the art world, allowing practically anybody with internet access to dream up competently-executed landscapes, portraits, sketches and comics — all without any talent whatsoever.

Critics say it’s the lowest common denominator of human expression, outsourcing to bloated algorithms that feasted on copyrighted materials while exploiting human artists who have yet to be fairly remunerated for having their life’s work be thrown into the AI wood chipper.

The raging debate has metastasized into a prolonged legal battle, with some attempting to uphold the legitimacy of AI-generated art by arguing it’s copyrightable — efforts that have met a major obstacle in the form of a recent Supreme Court decision.

In 2022, not long before text-to-image generative AI tools like Midjourney went mainstream, the US Copyright Office rejected computer scientist Stephen Thaler’s request to copyright his AI-generated image, titled “A Recent Entrance to Paradise.” After several years of back and forth, including an appeal, a US district court judge ruling that the work couldn’t be protected since it didn’t have a human creator, and eventually an affirming of said ruling in 2025, the case finally made it to the US Supreme Court.

And now, as Reuters reports, the country’s highest court has declined to hear the ongoing dispute, dealing a crushing blow to those who argue that AI-generated art should be eligible for copyright like human-created works.

It’s an especially thorny situation, considering AI companies are embroiled in a number of lawsuits over alleged copyright infringement of their own.

Image generator Midjourney, for instance, was sued by Warner Bros. Discovery last year. Artists also filed a lawsuit in 2024 against Google after finding their work had been scraped by the company’s AI. OpenAI’s ChatGPT and text-to-video generating app, Sora, can also easily be used to generate images and videos of copyrighted characters.

That hasn’t stopped a number of AI enthusiasts, including Thaler, from seeking copyright for their AI-generated work. Some of them have gone as far as to complain that their prompts are being plagiarized by other “artists.”

“Although the Copyright Act does not define the term ‘author,’ multiple provisions of the act make clear that the term refers to a human rather than a machine,” the Trump administration told Reuters in a statement.

Apart from copyrighting his artwork, Thaler also applied for patents for a food container and search and rescue beacon in 2018, arguing he wasn’t the inventor, but that an AI machine, dubbed DABUS, had come up with them.

The US Patent and Trademark Office shut him down, followed by the Supreme Court, which also denied hearing his argument, per Reuters.

More on AI art: Furious AI Users Say Their Prompts Are Being Plagiarized

The post The Supreme Court Just Dealt a Crushing Blow to “AI Artists” appeared first on Futurism.

🔗 Sumber: futurism.com