Improve dataset card: Add English summary, paper/project links, task category, and update metadata

README.md

@@ -1,59 +1,70 @@
 ---
 license: apache-2.0
+size_categories:
+- 10M<n<100M
+pretty_name: U2UData-2
 tags:
 - Swarm
 - UAVS
 - Autonoumous
 - Flying
 - Simulator
-pretty_name: U2USim-2
-size_categories:
-- 10K<n<100K
+- dataset
+task_categories:
+- robotics
 ---
 
-## 1. 文件分类
-- U2USim_2_Windows_version 仿真环境的windows打包版本
-- U2USim_2_Linux_version 仿真环境的Linux20.04打包版本
-- Settings.json 配置文件　放在\home\用户名\Documents\AirSim或者C:\Users\用户名\Documents\AirSim下 是对插件的一些预定义设置
-- AirDrone 基于API的算法控制、键盘控制、可视化交互控制器
-
-## 2. 系统需求
-### 2.1 PC需求
-- 内存至少32G以上
-- 显卡 RTX 4080，7900 XTX
-- CPU 主流CPU应该都行
-### 2.2 手柄 (经过测试的)
-- Xbox One， Xbox series X/S 的手柄都行，但是Xbox精英2代手柄不行
-- PS5 DuelSense controller 
-
-## 3. 运行环境需求和控制方式
+# U2UData-2: A Scalable Swarm UAVs Autonomous Flight Dataset for Long-horizon Tasks
+
+This repository is associated with the paper [U2UData-2: A Scalable Swarm UAVs Autonomous Flight Dataset for Long-horizon Tasks](https://huggingface.co/papers/2509.00055).
+
+Project page: [https://fengtt42.github.io/U2UData-2/](https://fengtt42.github.io/U2UData-2/)
+
+U2UData-2 is presented as the first large-scale swarm UAV autonomous flight dataset for Long-Horizon (LH) tasks and the first scalable swarm UAV data online collection and algorithm closed-loop verification platform. The dataset is captured by 15 UAVs in autonomous collaborative flights for LH tasks, comprising 12 scenes, 720 traces, 120 hours, 600 seconds per trajectory, 4.32 million LiDAR frames, and 12.96 million RGB frames. This dataset also includes brightness, temperature, humidity, smoke, and airflow values covering all flight routes. The platform supports the customization of simulators, UAVs, sensors, flight algorithms, formation modes, and LH tasks, allowing users to collect customized datasets and verify algorithms by closed-loop simulation.
+
+This repository provides components for the U2UData-2 platform and its associated simulator, U2USim-2.
+
+## 1. File Categories
+- `U2USim_2_Windows_version`: Windows packaged version of the simulation environment.
+- `U2USim_2_Linux_version`: Linux 20.04 packaged version of the simulation environment.
+- `Settings.json`: Configuration file, to be placed in `\\home\\username\\Documents\\AirSim` or `C:\\Users\\username\\Documents\\AirSim`. This file contains predefined settings for the plugin.
+- `AirDrone`: API-based algorithm control, keyboard control, and visual interactive controller.
+
+## 2. System Requirements
+### 2.1 PC Requirements
+- Memory: At least 32GB RAM.
+- Graphics Card: RTX 4080, 7900 XTX.
+- CPU: Mainstream CPUs should suffice.
+### 2.2 Gamepad (Tested)
+- Xbox One, Xbox Series X/S gamepads are compatible, but Xbox Elite Series 2 is not.
+- PS5 DualSense controller.
+
+## 3. Running Environment Requirements and Control Methods
 ### 3.1 U2USim_2_Windows_version
-- 不需要安装额外的运行库
-- 手柄控制 左手摇杆左右控制转向，上下飞机高度，右手控制飞机倾斜（移动），但是会让飞机下降
-- 仿真环境的键盘控制（在使用了EnableKeyboardControl之后无效）
-    - Backspace 飞机位置重置
-    - 键盘1234 打开图像可视化窗口
-    - F Fpv（第一人称）模式
-    - / 飞机固定在视角中下部
-    - End 重置整个仿真环境（重新打开关卡）
-    - R 录制传感器数据，鼠标点右下角红色按钮是同样的功能
+- No additional runtime libraries needed.
+- Gamepad controls: The left stick controls turning (left/right) and aircraft height (up/down). The right stick controls aircraft tilt (movement), but also causes descent.
+- Keyboard controls for simulation environment (invalid after `EnableKeyboardControl` is used):
+    - `Backspace`: Reset aircraft position.
+    - Keys `1`, `2`, `3`, `4`: Open image visualization window.
+    - `F`: FPV (First Person View) mode.
+    - `/`: Lock aircraft in the lower center of the view.
+    - `End`: Reset the entire simulation environment (reopen the level).
+    - `R`: Record sensor data (same function as clicking the red button in the bottom right).
 
 ### 3.2 U2USim_2_Linux_version
-- 控制方式和Windows相同
-- ROS信息获取参考 https://zhuanlan.zhihu.com/p/678127741
-- Airsim ROS package 仓库 https://github.com/seventyzlp/Codename-Aleph
+- Controls are the same as Windows.
+- For ROS information, refer to [https://zhuanlan.zhihu.com/p/678127741](https://zhuanlan.zhihu.com/p/678127741).
+- Airsim ROS package repository: [https://github.com/seventyzlp/Codename-Aleph](https://github.com/seventyzlp/Codename-Aleph).
 
 ### 3.3 AirDrone
-- 安装python运行库
-    - PyQt5
-    - keyboard
-    - numpy
-    - msgpack-rpc-python
-- 需要运行的主文件是AirDroneClient.py
-- 需要先打开仿真环境后，再启动这个控制器，不然会黑屏
-- 在点击EnableKeyboardControl按钮后，会强制接管所有键盘输入事件，关闭需要结束程序
-    - 键盘控制K是起飞，WASD 前后左右移动，上下 飞机高度升高降低， 左右控制飞机转向
-- 点击按键控制飞机不受影响
-- 在使用API控制飞机后，手柄的控制会被自动禁用，按下ExitApiControl按钮后可以重新使用手柄
-
-
+- Install Python libraries:
+    - `PyQt5`
+    - `keyboard`
+    - `numpy`
+    - `msgpack-rpc-python`
+- The main file to run is `AirDroneClient.py`.
+- The simulation environment must be opened before launching this controller, otherwise it will result in a black screen.
+- After clicking the `EnableKeyboardControl` button, all keyboard input events will be forcefully taken over. To disable, terminate the program.
+    - Keyboard controls: `K` for takeoff, `WASD` for forward/backward/left/right movement, Up/Down for increasing/decreasing aircraft height, Left/Right for turning.
+- Clicking buttons does not affect aircraft control.
+- After using API to control the aircraft, gamepad control will be automatically disabled. Gamepad can be re-enabled by pressing the `ExitApiControl` button.