MAROKO133 Breaking ai: Solar system’s asteroid belt slowly disappearing, houses over 1.9 m

📌 MAROKO133 Update ai: Solar system’s asteroid belt slowly disappearing, houses ov

Orbiting between Mars and Jupiter is an enormous ring of space rocks. The region, known as the asteroid belt, is home to an estimated 1.9 million asteroids that are larger than 1 kilometer in diameter. This makes up the vast majority of the solar system’s space rocks.

The belt, which marks the boundary between the inner rocky planets and the outer gas giants, is as old as the solar system itself. It formed from material that was present during the early formation of our star system’s planets.

Out of all the planets, Jupiter has the greatest impact on the asteroid belt. Due to its enormous size, the gas giant’s gravitational force occasionally pulls large space rocks out of the belt. That’s not the only impact it has, though. Jupiter’s gravitational pull also hurls enormous rocks around within the asteroid belt.

According to a new study, this effect is gradually pulverizing the asteroids that make up the belt. As they smash into each other, they are reduced to smaller pieces. As a result, the asteroid belt may be on course to completely disappear.

The asteroid belt’s slow demise

The new study, published in the preprint server arXiv, indicates that the asteroid belt is losing roughly 0.0088% of its mass every million years.

The team, led by planetary scientist Julio Fernández of the Universidad de la República in Uruguay, focused on the collisionally active portion of the asteroid belt. As a report from Gizmodo explains, this is the name given to the portion of the belt made up of asteroids small enough to be involved in frequent collisions and dynamical ejections.

According to Fernández and his colleagues’ calculations, the asteroid belt lost roughly a third of its mass over the past 3.5 billion years.

The team also estimates that 20 percent of asteroids escape into the inner and outer solar system. The other 80 percent, meanwhile, are simply pulverized and ground into cosmic dust within the asteroid belt. This dust finds its way into the zodiacal cloud – a thick dust cloud orbiting the Sun within the inner solar system.

Planetary defense implications

The scientists do point out that the death of the Sun, in about 5 billion years, will destroy the asteroid belt before it is able to fully vanish. This is because its rate of depletion is slowing over time, as fewer rocks lead to fewer collisions.

The findings are important, though, as they provide insight into the rate at which asteroids leave the asteroid belt. These could, of course, make their way toward our planet, meaning the data is important for planetary defense.

In 2022, NASA’s DART spacecraft smashed into an asteroid moonlet called Dimorphos in humanity’s first-ever planetary defense test. That test showed that we could alter an asteroid’s trajectory if one were found to be on a collision course with Earth.

Scientists have also hypothesized that a large portion of the world’s water comes from asteroids, and that the building blocks of life came to Earth on space rocks. The team’s data sheds new light on the universe’s ancient history, allowing scientists to better investigate the role asteroids have played in shaping our planet.

🔗 Sumber: interestingengineering.com

📌 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