Spaces:
Running
Running
File size: 12,640 Bytes
db561fa f27b3a7 5f1830f f27b3a7 5f1830f f27b3a7 db561fa | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 | <!doctype html>
<html lang="zh-CN">
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1">
<meta name="description" content="RabbitRobot VLN Lab: real-robot VLN/VLR for AMR, with deviation detection, verifiable recovery, ROS 2, Nav2, mapping and public robot assets.">
<title>RabbitRobot VLN Lab</title>
<style>
:root {
--ink: #172033;
--muted: #667085;
--line: #d8dee8;
--bg: #f4f7fb;
--blue: #1d4ed8;
--cyan: #0e7490;
--green: #168a55;
--orange: #e87022;
--paper: #ffffff;
--dark: #101827;
}
* { box-sizing: border-box; }
body {
margin: 0;
font-family: Inter, ui-sans-serif, system-ui, -apple-system, BlinkMacSystemFont, "Segoe UI", sans-serif;
color: var(--ink);
background: var(--bg);
letter-spacing: 0;
}
a { color: inherit; }
.hero {
min-height: 72vh;
display: grid;
align-items: end;
position: relative;
isolation: isolate;
color: #fff;
background-image:
linear-gradient(90deg, rgba(9, 18, 32, .88) 0%, rgba(9, 18, 32, .65) 44%, rgba(9, 18, 32, .18) 100%),
url("assets/rabbitrobot-amr-front.jpg");
background-size: cover;
background-position: center;
padding: clamp(28px, 6vw, 72px);
}
.hero-inner {
width: min(1060px, 100%);
display: grid;
grid-template-columns: minmax(0, 760px);
gap: 22px;
}
.kicker {
margin: 0;
text-transform: uppercase;
font-size: 13px;
font-weight: 800;
color: #ffcf99;
}
h1 {
margin: 0;
font-size: clamp(38px, 7vw, 78px);
line-height: .98;
letter-spacing: 0;
}
.hero p {
margin: 0;
max-width: 760px;
font-size: clamp(16px, 2vw, 22px);
line-height: 1.65;
color: #eef5ff;
}
.hero-actions {
display: flex;
flex-wrap: wrap;
gap: 10px;
}
.button {
display: inline-flex;
align-items: center;
justify-content: center;
min-height: 42px;
padding: 10px 16px;
border-radius: 7px;
border: 1px solid rgba(255,255,255,.34);
background: rgba(255,255,255,.12);
color: #fff;
text-decoration: none;
font-weight: 750;
}
.button.primary {
background: #fff;
color: #142033;
border-color: #fff;
}
.band {
padding: clamp(34px, 5vw, 64px) clamp(18px, 5vw, 72px);
}
.wrap {
width: min(1180px, 100%);
margin: 0 auto;
}
.section-head {
display: grid;
gap: 8px;
max-width: 860px;
margin-bottom: 22px;
}
h2 {
margin: 0;
font-size: clamp(24px, 3.2vw, 38px);
line-height: 1.15;
letter-spacing: 0;
}
.lead {
margin: 0;
color: var(--muted);
line-height: 1.75;
font-size: 16px;
}
.pipeline {
display: grid;
grid-template-columns: repeat(5, minmax(0, 1fr));
gap: 10px;
}
.step {
min-height: 172px;
padding: 16px;
border: 1px solid var(--line);
background: var(--paper);
border-radius: 8px;
display: grid;
align-content: start;
gap: 10px;
}
.step b {
color: var(--dark);
line-height: 1.25;
}
.step span {
color: var(--muted);
font-size: 14px;
line-height: 1.55;
}
.media-grid {
display: grid;
grid-template-columns: 1.15fr .85fr;
gap: 18px;
align-items: stretch;
}
.media {
margin: 0;
border: 1px solid var(--line);
border-radius: 8px;
overflow: hidden;
background: #fff;
}
.media img {
width: 100%;
height: 100%;
min-height: 320px;
object-fit: cover;
display: block;
}
.media figcaption {
padding: 12px 14px;
color: var(--muted);
font-size: 14px;
line-height: 1.55;
background: #fff;
}
.asset-grid {
display: grid;
grid-template-columns: repeat(3, minmax(0, 1fr));
gap: 14px;
}
.asset {
min-height: 230px;
display: grid;
gap: 12px;
align-content: start;
padding: 18px;
border: 1px solid var(--line);
border-radius: 8px;
background: var(--paper);
text-decoration: none;
}
.asset strong {
display: block;
color: var(--dark);
font-size: 18px;
line-height: 1.3;
}
.asset p {
margin: 0;
color: var(--muted);
line-height: 1.65;
font-size: 14px;
}
.tag-row {
display: flex;
flex-wrap: wrap;
gap: 7px;
margin-top: auto;
}
.tag {
display: inline-flex;
border-radius: 999px;
padding: 5px 8px;
background: #eff6ff;
color: #1d4ed8;
font-size: 12px;
font-weight: 750;
}
.dark-band {
background: #101827;
color: #f8fbff;
}
.dark-band .lead { color: #c6d3e5; }
.roadmap {
display: grid;
grid-template-columns: repeat(6, minmax(0, 1fr));
gap: 1px;
overflow: hidden;
border-radius: 8px;
border: 1px solid rgba(255,255,255,.14);
background: rgba(255,255,255,.14);
}
.roadmap div {
min-height: 112px;
padding: 14px;
background: #172238;
color: #dce8f7;
font-size: 14px;
line-height: 1.5;
}
.roadmap b {
display: block;
color: #fff;
margin-bottom: 6px;
}
.closing {
display: grid;
grid-template-columns: 1fr auto;
gap: 20px;
align-items: center;
border-top: 1px solid var(--line);
}
.closing p {
margin: 0;
color: var(--muted);
line-height: 1.7;
}
.closing .button {
color: #fff;
background: var(--dark);
border-color: var(--dark);
}
@media (max-width: 980px) {
.pipeline,
.asset-grid,
.roadmap {
grid-template-columns: repeat(2, minmax(0, 1fr));
}
.media-grid,
.closing {
grid-template-columns: 1fr;
}
}
@media (max-width: 620px) {
.hero {
min-height: 68vh;
padding: 26px 18px;
background-position: 62% center;
}
.pipeline,
.asset-grid,
.roadmap {
grid-template-columns: 1fr;
}
.button {
width: 100%;
}
}
</style>
</head>
<body>
<header class="hero">
<div class="hero-inner">
<p class="kicker">RabbitRobot VLN / VLR Lab</p>
<h1>真实 AMR 上的长程导航与可验证恢复</h1>
<p>RabbitRobot 是我的真实机器人研究平台:把 ROS 2、Nav2、地图、传感器、Web 工具和公开数据资产连接起来,服务 Vision-Language Navigation / Vision-Language Recovery。</p>
<div class="hero-actions">
<a class="button primary" href="https://huggingface.co/datasets/lijinghai/RabbitRobot-VLN-Recovery-Mini">Open VLN Dataset</a>
<a class="button" href="https://huggingface.co/collections/lijinghai/rabbitrobot-vln-and-open-robot-assets-6a71a24e6d087ac0ff031ef5">Open Collection</a>
</div>
</div>
</header>
<main>
<section class="band">
<div class="wrap">
<div class="section-head">
<p class="kicker">Research Spine</p>
<h2>VLN 不只是听懂指令,而是把语言目标落到真实机器人执行闭环里。</h2>
<p class="lead">当前主线关注长程 VLN 中的偏航检测、目标歧义处理、语义子目标规划、任务记忆保持和可验证恢复。公开资产会先从小而干净的数据结构开始,逐步扩展到 rosbag replay、视觉 grounding 和真实 AMR 评估。</p>
</div>
<div class="pipeline">
<div class="step"><b>Language Goal</b><span>自然语言目标、空间关系、语义子目标和任务阶段记忆。</span></div>
<div class="step"><b>Grounding</b><span>把语言目标落到地图、视觉帧、waypoint 候选和可到达区域。</span></div>
<div class="step"><b>Nav2 Execution</b><span>路径通过 ROS 2 / Nav2 / costmap / controller,而不是只画规划线。</span></div>
<div class="step"><b>Deviation Detection</b><span>用里程计、costmap、局部规划状态和视觉信号发现偏航。</span></div>
<div class="step"><b>Verifiable Recovery</b><span>恢复到安全、可解释、可继续执行的状态,并记录评估证据。</span></div>
</div>
</div>
</section>
<section class="band">
<div class="wrap media-grid">
<figure class="media">
<img src="assets/amr2-web-navigation.jpg" alt="RabbitRobot AMR web navigation interface">
<figcaption>RabbitRobot 的导航与 Web 工具链:真实机器人状态、地图和任务入口需要能被观察、复盘和继续开发。</figcaption>
</figure>
<figure class="media">
<img src="assets/coverage-generated-path.png" alt="RabbitRobot coverage path planning overlay">
<figcaption>覆盖路径规划是 VLN 执行侧的工程支撑:区域选择、known-free 裁剪、车体安全膨胀和 waypoint 生成。</figcaption>
</figure>
</div>
</section>
<section class="band">
<div class="wrap">
<div class="section-head">
<p class="kicker">Open Assets</p>
<h2>当前公开资产按 VLN 主线组织。</h2>
<p class="lead">先发布可公开、可解释、可继续扩展的小资产;真实大数据会在脱敏、切片和文档完善后逐步进入。</p>
</div>
<div class="asset-grid">
<a class="asset" href="https://huggingface.co/datasets/lijinghai/RabbitRobot-VLN-Recovery-Mini">
<strong>RabbitRobot VLN Recovery Mini</strong>
<p>长程 VLN/VLR task cards:自然语言指令、偏航事件、可观测信号、恢复策略和成功标准。</p>
<div class="tag-row"><span class="tag">VLN</span><span class="tag">Recovery</span><span class="tag">Dataset</span></div>
</a>
<a class="asset" href="https://huggingface.co/spaces/lijinghai/RabbitRobot-CoveragePath-Demo">
<strong>RabbitRobot Coverage Path Planning</strong>
<p>静态交互演示:地图约束、footprint clearance、known-free clipping 与覆盖 waypoint 生成。</p>
<div class="tag-row"><span class="tag">Nav2</span><span class="tag">Planning</span><span class="tag">Space</span></div>
</a>
<a class="asset" href="https://huggingface.co/datasets/lijinghai/R550-ROS2-Graph-Dictionary">
<strong>R550 ROS2 Graph Dictionary</strong>
<p>真实机器人 ROS 2 图谱字典:节点、Topic、Service、命令和功能域关系,服务系统理解与调试。</p>
<div class="tag-row"><span class="tag">ROS 2</span><span class="tag">Graph</span><span class="tag">Dataset</span></div>
</a>
</div>
</div>
</section>
<section class="band dark-band">
<div class="wrap">
<div class="section-head">
<p class="kicker">Roadmap</p>
<h2>从真实机器人底座走向可复现 VLN 评估。</h2>
<p class="lead">这条线的价值不是单次演示,而是让机器人系统、实验数据和评估协议逐步变成别人可以检查、复现和继续研究的开放资产。</p>
</div>
<div class="roadmap">
<div><b>1. AMR Baseline</b>ROS 2 / Nav2 / mapping / sensor integration.</div>
<div><b>2. System Graph</b>ROS 2 nodes, topics, services and launch flows.</div>
<div><b>3. Planning Demo</b>Coverage path and safe execution-chain thinking.</div>
<div><b>4. VLN Cards</b>Instruction grounding, deviation events and recovery labels.</div>
<div><b>5. Replay</b>Filtered rosbag slices, visual grounding and failure traces.</div>
<div><b>6. Evaluation</b>Success, deviation, recovery and explainability metrics.</div>
</div>
</div>
</section>
<section class="band closing">
<div>
<h2>面向猎头、导师和合作者</h2>
<p>如果你想快速判断我的方向:我是把真实机器人系统、ROS 2 工程和 VLN/VLR 研究问题放在同一条线上做的人。更完整的项目经历在个人主页和 GitHub。</p>
</div>
<a class="button" href="https://lijinghai.github.io/">Open Personal Site</a>
</section>
</main>
</body>
</html>
|