536 lines
33 KiB
HTML
536 lines
33 KiB
HTML
<!DOCTYPE html>
|
||
<html lang="zh-CN" data-theme="night">
|
||
<head>
|
||
<meta charset="UTF-8">
|
||
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
||
<title>机器人动作捕捉 - 深度行研报告</title>
|
||
<script src="https://cdn.tailwindcss.com"></script>
|
||
<link href="https://cdn.jsdelivr.net/npm/daisyui@4.11.1/dist/full.min.css" rel="stylesheet" type="text/css" />
|
||
<script defer src="https://cdn.jsdelivr.net/npm/alpinejs@3.x.x/dist/cdn.min.js"></script>
|
||
<script src="https://cdn.jsdelivr.net/npm/echarts@5.5.0/dist/echarts.min.js"></script>
|
||
<style>
|
||
@import url('https://fonts.googleapis.com/css2?family=Oxanium:wght@300;400;500;700&family=Noto+Sans+SC:wght@300;400;500;700&display=swap');
|
||
|
||
body {
|
||
font-family: 'Noto Sans SC', sans-serif;
|
||
background-color: #00000a;
|
||
color: #e0e0e0;
|
||
overflow-x: hidden;
|
||
}
|
||
|
||
.font-fui {
|
||
font-family: 'Oxanium', cursive;
|
||
}
|
||
|
||
.glass-card {
|
||
background: rgba(15, 18, 48, 0.4); /* Deeper blue tint */
|
||
backdrop-filter: blur(20px);
|
||
-webkit-backdrop-filter: blur(20px);
|
||
border: 1px solid rgba(118, 142, 255, 0.15);
|
||
border-radius: 1.5rem; /* 24px */
|
||
transition: all 0.3s ease;
|
||
}
|
||
|
||
.glass-card:hover {
|
||
border-color: rgba(132, 161, 255, 0.4);
|
||
box-shadow: 0 0 30px rgba(100, 130, 255, 0.2);
|
||
}
|
||
|
||
.glow-bg {
|
||
position: fixed;
|
||
top: 0;
|
||
left: 0;
|
||
width: 100%;
|
||
height: 100%;
|
||
z-index: -10;
|
||
overflow: hidden;
|
||
background-color: #00000a;
|
||
}
|
||
|
||
.glow-bg::before, .glow-bg::after {
|
||
content: '';
|
||
position: absolute;
|
||
border-radius: 50%;
|
||
filter: blur(150px);
|
||
opacity: 0.2;
|
||
}
|
||
|
||
.glow-bg::before {
|
||
background: radial-gradient(circle, #4a00e0, transparent 60%);
|
||
width: 800px;
|
||
height: 800px;
|
||
top: -20%;
|
||
left: -20%;
|
||
animation: pulse-glow-1 15s infinite alternate;
|
||
}
|
||
|
||
.glow-bg::after {
|
||
background: radial-gradient(circle, #00bfff, transparent 60%);
|
||
width: 600px;
|
||
height: 600px;
|
||
bottom: -30%;
|
||
right: -25%;
|
||
animation: pulse-glow-2 18s infinite alternate;
|
||
}
|
||
|
||
@keyframes pulse-glow-1 {
|
||
from { transform: scale(0.8) translate(50px, -50px); opacity: 0.15; }
|
||
to { transform: scale(1.2) translate(-50px, 50px); opacity: 0.25; }
|
||
}
|
||
|
||
@keyframes pulse-glow-2 {
|
||
from { transform: scale(1.2) translate(-30px, 30px); opacity: 0.25; }
|
||
to { transform: scale(0.9) translate(30px, -30px); opacity: 0.15; }
|
||
}
|
||
|
||
.section-title {
|
||
@apply text-3xl font-fui font-bold text-transparent bg-clip-text bg-gradient-to-r from-cyan-300 to-indigo-400 mb-6;
|
||
}
|
||
.subsection-title {
|
||
@apply text-xl font-fui font-semibold text-cyan-200 mt-6 mb-3 border-l-4 border-cyan-400 pl-4;
|
||
}
|
||
.insight-section h3 {
|
||
@apply text-xl font-fui font-semibold text-cyan-200 mt-6 mb-3 border-l-4 border-cyan-400 pl-4;
|
||
}
|
||
|
||
.table thead th {
|
||
background-color: rgba(132, 161, 255, 0.1);
|
||
color: #c0d0ff;
|
||
font-family: 'Oxanium', cursive;
|
||
font-size: 1rem;
|
||
}
|
||
.table tbody tr {
|
||
border-bottom: 1px solid rgba(118, 142, 255, 0.1);
|
||
transition: background-color 0.3s;
|
||
}
|
||
.table tbody tr:hover {
|
||
background-color: rgba(132, 161, 255, 0.08);
|
||
}
|
||
|
||
.tab {
|
||
transition: all 0.3s ease;
|
||
color: #a0b0ff;
|
||
}
|
||
|
||
.tab-active {
|
||
color: #ffffff;
|
||
font-weight: bold;
|
||
border-bottom: 2px solid #6366f1;
|
||
}
|
||
</style>
|
||
</head>
|
||
<body class="min-h-screen">
|
||
|
||
<div class="glow-bg"></div>
|
||
|
||
<div class="container mx-auto p-4 sm:p-6 md:p-8">
|
||
|
||
<header class="text-center mb-12 animate-fade-in-down">
|
||
<h1 class="text-5xl md:text-6xl font-fui font-extrabold text-transparent bg-clip-text bg-gradient-to-r from-sky-300 via-indigo-400 to-purple-500 py-2">
|
||
机器人动作捕捉
|
||
</h1>
|
||
<p class="mt-4 text-xl text-indigo-200 font-light">深度行研报告</p>
|
||
<div class="mt-4 text-xs text-slate-500">
|
||
<p>北京价值前沿科技有限公司 AI投研agent:“价小前投研” 进行投研呈现</p>
|
||
<p>本报告为AI合成数据,不构成任何投资建议,投资需谨慎。</p>
|
||
</div>
|
||
</header>
|
||
|
||
<main class="space-y-12">
|
||
<!-- Core Insight Summary -->
|
||
<section class="glass-card p-8">
|
||
<h2 class="section-title">核心观点摘要</h2>
|
||
<p class="text-lg text-slate-300 leading-relaxed">
|
||
机器人动作捕捉正从影视、游戏等传统领域的“可选工具”,迅速蜕变为人形机器人实现高级智能与泛化能力的“<strong class="text-cyan-300 font-semibold">核心基础设施</strong>”。其核心驱动力在于解决人形机器人量产前最关键的<strong class="text-cyan-300 font-semibold">高质量、低成本训练数据</strong>的供给瓶颈。目前,该概念正处于从“<strong class="text-indigo-300">研发采购驱动</strong>”向“<strong class="text-indigo-300">数据服务驱动</strong>”演进的早期阶段,市场空间巨大,但技术路线与商业模式仍在探索与验证中。
|
||
</p>
|
||
</section>
|
||
|
||
<!-- Bento Grid Overview -->
|
||
<section class="grid md:grid-cols-3 gap-6">
|
||
<div class="glass-card p-6 md:col-span-2">
|
||
<h3 class="subsection-title">核心驱动力</h3>
|
||
<ul class="list-disc list-inside space-y-2 text-slate-300">
|
||
<li><strong class="text-indigo-300">数据稀缺性:</strong>人形机器人泛化能力需要百万小时级动作数据,动捕是兼具精度和效率的最佳来源。</li>
|
||
<li><strong class="text-indigo-300">技术路径必然选择:</strong>行业已验证“动捕数据 → 仿真训练 → 实体复现”路径的有效性,如宇树科技案例。</li>
|
||
<li><strong class="text-indigo-300">行业标杆示范效应:</strong>特斯拉CEO马斯克公开强调动捕重要性,并已部署24小时不间断数据采集团队。</li>
|
||
<li><strong class="text-indigo-300">成本下降经济驱动:</strong>数据采集成本有望在三年内从每秒300元降至百元级,提升经济可行性。</li>
|
||
</ul>
|
||
</div>
|
||
<div class="glass-card p-6 flex flex-col justify-center">
|
||
<h3 class="subsection-title">市场热度与情绪</h3>
|
||
<p class="text-slate-300">当前市场关注度极高,情绪整体偏向<strong class="text-green-400">乐观</strong>。研报密集发布、新闻热度持续、关联个股活跃,市场普遍认为动捕将迎来类似AI服务器的“卖铲子”行情。</p>
|
||
</div>
|
||
</section>
|
||
|
||
<!-- Technology & Market Chart -->
|
||
<section class="grid lg:grid-cols-5 gap-6">
|
||
<div class="lg:col-span-3 glass-card p-6">
|
||
<h3 class="subsection-title">技术路线对比</h3>
|
||
<div id="tech-comparison-chart" style="width: 100%; height: 400px;"></div>
|
||
</div>
|
||
<div class="lg:col-span-2 glass-card p-6">
|
||
<h3 class="subsection-title">高端市场格局</h3>
|
||
<div id="market-share-chart" style="width: 100%; height: 400px;"></div>
|
||
</div>
|
||
</section>
|
||
|
||
<!-- Data Source Tabs -->
|
||
<section class="glass-card p-6" x-data="{ tab: 'news' }">
|
||
<div role="tablist" class="tabs tabs-bordered mb-6">
|
||
<a role="tab" class="tab" :class="{ 'tab-active': tab === 'news' }" @click.prevent="tab = 'news'">新闻数据</a>
|
||
<a role="tab" class="tab" :class="{ 'tab-active': tab === 'roadshow' }" @click.prevent="tab = 'roadshow'">路演纪要</a>
|
||
<a role="tab" class="tab" :class="{ 'tab-active': tab === 'research' }" @click.prevent="tab = 'research'">研报精粹</a>
|
||
</div>
|
||
|
||
<div x-show="tab === 'news'" x-transition>
|
||
<div class="prose max-w-none prose-invert prose-p:text-slate-300 prose-li:text-slate-300 prose-strong:text-indigo-300">
|
||
<h4>市场趋势与核心观点</h4>
|
||
<ul>
|
||
<li><strong>需求增长:</strong> 机器人领域动捕训练需求增长,是人形机器人动作迈向高度拟人化的重要推手。</li>
|
||
<li><strong>数据为王:</strong> 低成本采集高质量数据是量产前急需解决的问题,动捕采集的数据精度最高、能力最全面。</li>
|
||
<li><strong>成本下降:</strong> 数据采集成本三年内或从每秒300元降至百元级。</li>
|
||
<li><strong>行业关注:</strong> 宇树机器人格斗大赛超预期,动捕迎来高度关注。</li>
|
||
</ul>
|
||
<h4>公司与产品应用</h4>
|
||
<ul>
|
||
<li><strong>特斯拉:</strong> 率先采用百万级动捕设备构建虚实双模数据工厂。</li>
|
||
<li><strong>利亚德/虚拟动点:</strong> 以空间计算为基础入局,定位为“机器人的大脑和眼睛”,提供数据开发+机器人训练服务。已发布利亚德动作大模型,对接不低于十个机器人客户。</li>
|
||
<li><strong>凌云光:</strong> 国内自主龙头,FZMotion系统提供全链条解决方案,国内领先机器人厂商几乎全覆盖,已在宇树等头部厂商批量出货。</li>
|
||
<li><strong>宇树科技:</strong> 动作仿真性能优越,主要得益于光学动作捕捉+仿真学习。</li>
|
||
</ul>
|
||
<h4>关键技术与专利</h4>
|
||
<ul>
|
||
<li><strong>技术路线:</strong> 主流为光学、光惯混合。虚拟动点主推光学与无标记技术融合。</li>
|
||
<li><strong>核心算法:</strong> 亚毫米级动捕精度通过“人体-机械轴映射算法”突破产业瓶颈;利亚德动作大模型LYDIA已接入DeepSeek。</li>
|
||
<li><strong>宇树专利:</strong> “一种基于动捕设备的机器人关节控制方法和系统”获授权,可高效准确地将人类动作映射至机器人。</li>
|
||
</ul>
|
||
</div>
|
||
</div>
|
||
<div x-show="tab === 'roadshow'" x-transition>
|
||
<div class="prose max-w-none prose-invert prose-p:text-slate-300 prose-li:text-slate-300 prose-strong:text-indigo-300">
|
||
<h4>应用与目的</h4>
|
||
<ul>
|
||
<li><strong>核心目的:</strong> 为机器人训练提供高质量动作数据,是必须环节。</li>
|
||
<li><strong>技术路径:</strong> 主要通过模仿学习(Imitation Learning)和强化学习(Reinforcement Learning)来训练。</li>
|
||
<li><strong>流程:</strong> 动捕设备采集 → 数据传入仿真软件训练虚拟本体 → 迁移至实体机器人。</li>
|
||
</ul>
|
||
<h4>技术路线对比</h4>
|
||
<ul>
|
||
<li><strong>光学动捕:</strong> 精度高 (亚毫米级),适合复杂动作,但需固定大场地,成本高。国内99%厂商使用。</li>
|
||
<li><strong>惯性动捕 (IMU):</strong> 便携灵活,不受场地光线限制,但存在漂移问题,精度低于光学。特斯拉擎天柱项目采用此技术。</li>
|
||
<li><strong>路线预测:</strong> 光学方案仍占主导(70-80%),但惯性方案潜力被低估,尤其在服务机器人领域。</li>
|
||
</ul>
|
||
<h4>市场格局与成本</h4>
|
||
<ul>
|
||
<li><strong>高端市场:</strong> Vicon(英国)与OptiTrack(美国)垄断,合计占全球80%以上份额。</li>
|
||
<li><strong>国内厂商:</strong> 光学(度量科技、凌云光等)、惯性(诺亦腾)。</li>
|
||
<li><strong>成本:</strong> 高端惯性系统(Xsens)约30-41万/套;光学系统(度量科技)约30-50万/套;自建50相机级光学棚约150-300万。</li>
|
||
</ul>
|
||
<h4>商业模式与趋势</h4>
|
||
<ul>
|
||
<li><strong>当前模式:</strong> 本体厂商采购少量硬件自用。</li>
|
||
<li><strong>长期趋势:</strong> 轻资产模式崛起,本体厂商转向购买数据服务,动捕厂商向“设备+数据”服务商转型。</li>
|
||
</ul>
|
||
</div>
|
||
</div>
|
||
<div x-show="tab === 'research'" x-transition>
|
||
<div class="prose max-w-none prose-invert prose-p:text-slate-300 prose-li:text-slate-300 prose-strong:text-indigo-300">
|
||
<h4>核心地位与重要性</h4>
|
||
<ul>
|
||
<li><strong>破局关键:</strong> 动捕系统是具身智能数据采集的“破局关键手段”,从“可选工具”向“核心基础设施”跃迁。</li>
|
||
<li><strong>数据质量保证:</strong> 泛化能力提升需要海量高精度数据,动捕能直接决定数据集质量。</li>
|
||
<li><strong>政策支持:</strong> 北京、上海等地明确提出要“开展具身智能数据采集,开放动作数据集”。</li>
|
||
</ul>
|
||
<h4>技术分类与壁垒</h4>
|
||
<ul>
|
||
<li><strong>主流技术:</strong> 光学式(高精度)和惯性式(便携)。研报认为惯性动捕更适合机器人运动(平衡、跌倒检测)。</li>
|
||
<li><strong>技术壁垒:</strong> 高精度传感器、抗磁干扰、高刷新率及核心的本体映射、平台适配算法。</li>
|
||
</ul>
|
||
<h4>市场前景与规模</h4>
|
||
<ul>
|
||
<li><strong>当前市场:</strong> 2022年市场空间仅58亿元,下游以电影、游戏为主。</li>
|
||
<li><strong>未来增量:</strong> 假设单个人形机器人企业使用1000台训练,有望为动捕市场带来<strong>500亿</strong>增量空间。</li>
|
||
</ul>
|
||
<h4>主要参与者</h4>
|
||
<ul>
|
||
<li><strong>Movella (Xsens):</strong> 全球领先,特斯拉采用其方案。</li>
|
||
<li><strong>诺亦腾 (Noitom):</strong> 奥飞娱乐投资,布局惯性技术,与NVIDIA Isaac合作。</li>
|
||
<li><strong>凌云光:</strong> 全资子公司元客视界推出FZmotion,服务宇树、优必选等。</li>
|
||
<li><strong>利亚德 (虚拟动点):</strong> 拥有OptiTrack光学产品和高品质数据,与松延动力合作。</li>
|
||
</ul>
|
||
</div>
|
||
</div>
|
||
</section>
|
||
|
||
<!-- Full Insight Analysis -->
|
||
<section class="glass-card p-8 insight-section">
|
||
<h2 class="section-title">综合Insight深度剖析</h2>
|
||
|
||
<h3>概念事件与发展脉络</h3>
|
||
<p class="text-slate-300">“机器人动作捕捉”概念由技术需求、行业实践与市场预期共同催化。特斯拉、波士顿动力的长期应用奠定技术基础;宇树科技格斗大赛的惊艳表现引爆市场认知;利亚德、凌云光等国内厂商的业务倾斜与批量出货,以及京沪等地的政策加持,共同推动产业化预期升温。</p>
|
||
|
||
<h3>核心逻辑与市场认知分析</h3>
|
||
<p class="text-slate-300">核心逻辑围绕“数据”展开,动捕是解决人形机器人高质量训练数据稀缺性的最佳方案。市场热度高涨,普遍预期“卖铲子”行情。然而,存在关键预期差:</p>
|
||
<ul class="list-decimal list-inside space-y-2 mt-4 text-slate-300 pl-4">
|
||
<li><strong class="text-cyan-300">技术路线认知偏差:</strong>市场普遍认为光学动捕是绝对主流,但严重低估了行业风向标<strong class="text-yellow-300">特斯拉采用惯性动捕方案</strong>的战略意义。惯性方案在特定场景下具备独特优势,市场可能过度押注光学路线。</li>
|
||
<li><strong class="text-cyan-300">商业模式认知滞后:</strong>市场关注点多在设备销售,但核心厂商已向“设备+数据服务(DaaS)”转型,这种模式具备更高客户粘性和利润空间,其价值尚未被市场充分认知。</li>
|
||
<li><strong class="text-cyan-300">替代技术潜在威胁:</strong>长期来看,基于AI从2D视频反算3D数据的纯视觉方案若取得突破,可能颠覆现有格局,这是市场乐观情绪中潜藏的远期风险。</li>
|
||
</ul>
|
||
|
||
<h3>关键催化剂与未来发展路径</h3>
|
||
<p class="text-slate-300"><strong>近期催化剂 (3-6个月):</strong> 头部机器人厂商的大规模采购订单、标志性数据服务合同落地、行业级动作数据集的发布。</p>
|
||
<p class="text-slate-300"><strong>长期发展路径:</strong></p>
|
||
<ol class="list-decimal list-inside space-y-2 mt-4 text-slate-300 pl-4">
|
||
<li><strong>第一阶段 (当前-2026年): 硬件采购潮。</strong>设备商迎来业绩爆发。</li>
|
||
<li><strong>第二阶段 (2026-2028年): 服务模式兴起。</strong>第三方数据服务商崛起。</li>
|
||
<li><strong>第三阶段 (2028年以后): 生态与平台化。</strong>平台型企业诞生,并可能面临纯视觉方案的竞争。</li>
|
||
</ol>
|
||
|
||
<h3>产业链与核心公司深度剖析</h3>
|
||
<div class="overflow-x-auto mt-4">
|
||
<table class="table table-zebra w-full">
|
||
<thead>
|
||
<tr>
|
||
<th>公司</th>
|
||
<th>代码</th>
|
||
<th>核心优势</th>
|
||
<th>潜在风险</th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr>
|
||
<td class="font-semibold text-white">凌云光</td>
|
||
<td>688400</td>
|
||
<td class="text-green-300">国内光学动捕自主龙头,全链条解决方案,客户绑定深,逻辑最纯粹,业绩兑现最快。</td>
|
||
<td class="text-amber-300">过度依赖国内市场,技术路线单一(主攻光学)。</td>
|
||
</tr>
|
||
<tr>
|
||
<td class="font-semibold text-white">利亚德</td>
|
||
<td>300296</td>
|
||
<td class="text-green-300">持有全球顶尖品牌OptiTrack,率先转型“数据服务”,发布动作大模型,想象空间大。</td>
|
||
<td class="text-amber-300">机器人业务尚在起步,对整体业绩贡献有限,需耐心等待放量。</td>
|
||
</tr>
|
||
<tr>
|
||
<td class="font-semibold text-white">奥飞娱乐</td>
|
||
<td>002292</td>
|
||
<td class="text-green-300">通过诺亦腾卡位<strong class="text-yellow-400">惯性动捕</strong>赛道(特斯拉路线),存在预期差,与NVIDIA生态联动。</td>
|
||
<td class="text-amber-300">动捕非主业,持股比例有限(5%),业绩弹性相对较小。</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
</div>
|
||
|
||
<h3>潜在风险与挑战</h3>
|
||
<ul class="list-disc list-inside space-y-2 mt-4 text-slate-300">
|
||
<li><strong>技术风险:</strong> “人体-机械轴映射算法”仍是瓶颈;惯性技术存在漂移和电磁干扰问题;长期面临纯视觉方案的替代冲击。</li>
|
||
<li><strong>商业化风险:</strong> 设备成本高昂可能延缓中小企业研发;最终需求与人形机器人量产进度强绑定。</li>
|
||
<li><strong>信息交叉验证风险:</strong> 最显著的矛盾点在于,国内厂商实践(宇树)和市场普遍认知倾向<strong class="text-cyan-400">光学</strong>,而全球领军者特斯拉却采用<strong class="text-yellow-400">惯性</strong>方案。这揭示了市场可能存在的认知盲区,不同应用场景催生不同技术偏好,而非单一技术独大。</li>
|
||
</ul>
|
||
|
||
<h3>综合结论与投资启示</h3>
|
||
<p class="text-slate-300"><strong>阶段判断:</strong>概念已走出纯主题炒作,进入由研发需求驱动的<strong class="text-lime-300">早期基本面兑现阶段</strong>。</p>
|
||
<p class="text-slate-300"><strong>投资价值排序:</strong></p>
|
||
<ol class="list-decimal list-inside space-y-2 mt-4 text-slate-300 pl-4">
|
||
<li><strong class="text-sky-300">“设备+数据服务”双轮驱动的平台型公司(如利亚德):</strong>最具长期价值。</li>
|
||
<li><strong class="text-sky-300">卡位惯性动捕赛道的公司(如奥飞娱乐):</strong>最具预期差。</li>
|
||
<li><strong class="text-sky-300">业绩兑现最快的纯硬件龙头(如凌云光):</strong>最具短期确定性。</li>
|
||
</ol>
|
||
<p class="text-slate-300 mt-4"><strong>需重点跟踪的关键指标:</strong>下游渗透率、订单质量、<strong class="text-yellow-300">服务收入占比</strong>、以及新发布机器人项目采用光学/惯性方案的比例。</li>
|
||
|
||
</section>
|
||
|
||
<!-- Stock List Table -->
|
||
<section class="glass-card p-6">
|
||
<h2 class="section-title">相关概念股梳理</h2>
|
||
<div class="overflow-x-auto">
|
||
<table class="table w-full">
|
||
<thead>
|
||
<tr>
|
||
<th>股票名称</th>
|
||
<th>股票代码</th>
|
||
<th>核心逻辑</th>
|
||
<th>分类标签</th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr>
|
||
<td class="font-semibold text-white">凌云光</td>
|
||
<td><a href="https://valuefrontier.cn/company?scode=688400" target="_blank" class="link link-hover text-indigo-300">688400</a></td>
|
||
<td>自研FZMotion光学系统(亚毫米级精度),支持机器人训练、影视动画、工业仿真,服务优必选、小米等客户</td>
|
||
<td><span class="badge badge-primary">光学动捕</span> <span class="badge badge-accent">核心标的</span></td>
|
||
</tr>
|
||
<tr>
|
||
<td class="font-semibold text-white">利亚德</td>
|
||
<td><a href="https://valuefrontier.cn/company?scode=300296" target="_blank" class="link link-hover text-indigo-300">300296</a></td>
|
||
<td>子公司虚拟动点,拥有OptiTrack光学动捕产品及动作数据库,与松延动力合作成立“具身智能机器人联合实验室”</td>
|
||
<td><span class="badge badge-primary">光学动捕</span> <span class="badge badge-accent">核心标的</span></td>
|
||
</tr>
|
||
<tr>
|
||
<td class="font-semibold text-white">奥飞娱乐</td>
|
||
<td><a href="https://valuefrontier.cn/company?scode=002292" target="_blank" class="link link-hover text-indigo-300">002292</a></td>
|
||
<td>参股的诺亦腾,提供惯性+光学混合方案,与NVIDIA Isaac平台数据联通,客户案例包括特斯拉、智元机器人</td>
|
||
<td><span class="badge badge-secondary">惯性动捕</span> <span class="badge badge-accent">核心标的</span></td>
|
||
</tr>
|
||
<tr>
|
||
<td class="font-semibold text-white">捷成股份</td>
|
||
<td><a href="https://valuefrontier.cn/company?scode=300182" target="_blank" class="link link-hover text-indigo-300">300182</a></td>
|
||
<td>参股世优科技:惯导动捕技术应用于数字人领域,与北京大学共建实验室,积累数据库</td>
|
||
<td><span class="badge badge-secondary">惯性动捕</span></td>
|
||
</tr>
|
||
<tr>
|
||
<td class="font-semibold text-white">华依科技</td>
|
||
<td><a href="https://valuefrontier.cn/company?scode=688071" target="_blank" class="link link-hover text-indigo-300">688071</a></td>
|
||
<td>在多模态训练中具备卡位优势,与凌云光合作推进机器人感知层技术</td>
|
||
<td><span class="badge badge-ghost">产业链关联</span></td>
|
||
</tr>
|
||
<tr>
|
||
<td class="font-semibold text-white">天娱数科</td>
|
||
<td><a href="https://valuefrontier.cn/company?scode=002354" target="_blank" class="link link-hover text-indigo-300">002354</a></td>
|
||
<td>通过投资银牛微电子、芯明智能布局动捕硬件,构建多模态数据集,应用于数字人及机器人训练</td>
|
||
<td><span class="badge badge-ghost">产业链关联</span></td>
|
||
</tr>
|
||
<tr>
|
||
<td class="font-semibold text-white">瑞立视</td>
|
||
<td>-</td>
|
||
<td>对标Vicon,主攻工业机器人追踪及校准</td>
|
||
<td><span class="badge badge-primary">光学动捕</span> <span class="badge badge-info">未上市</span></td>
|
||
</tr>
|
||
<tr>
|
||
<td class="font-semibold text-white">青瞳视觉</td>
|
||
<td>-</td>
|
||
<td>主打低延迟红外系统,应用于机器人动作测试及影视制作</td>
|
||
<td><span class="badge badge-secondary">惯性动捕</span> <span class="badge badge-info">未上市</span></td>
|
||
</tr>
|
||
<tr>
|
||
<td class="font-semibold text-white">度量科技</td>
|
||
<td>-</td>
|
||
<td>光学动捕系统,覆盖高校科研市场</td>
|
||
<td><span class="badge badge-primary">光学动捕</span> <span class="badge badge-info">未上市</span></td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
</div>
|
||
</section>
|
||
</main>
|
||
</div>
|
||
|
||
<script>
|
||
document.addEventListener('DOMContentLoaded', function () {
|
||
// ECharts for Tech Comparison
|
||
var techChartDom = document.getElementById('tech-comparison-chart');
|
||
var techChart = echarts.init(techChartDom, 'dark');
|
||
var techOption = {
|
||
backgroundColor: 'transparent',
|
||
color: ['#00BFFF', '#8A2BE2'],
|
||
tooltip: {
|
||
trigger: 'item'
|
||
},
|
||
legend: {
|
||
data: ['光学动捕', '惯性动捕'],
|
||
textStyle: { color: '#ccc' }
|
||
},
|
||
radar: {
|
||
indicator: [
|
||
{ name: '精度', max: 100 },
|
||
{ name: '灵活性', max: 100 },
|
||
{ name: '抗干扰', max: 100 },
|
||
{ name: '成本', max: 100 },
|
||
{ name: '场地要求', max: 100 },
|
||
],
|
||
axisName: {
|
||
color: '#e0e0e0',
|
||
fontSize: 14,
|
||
},
|
||
splitArea: {
|
||
areaStyle: {
|
||
color: ['rgba(50, 50, 100, 0.2)', 'rgba(50, 50, 100, 0.4)'],
|
||
shadowColor: 'rgba(0, 0, 0, 0.2)',
|
||
shadowBlur: 10
|
||
}
|
||
},
|
||
axisLine: { lineStyle: { color: 'rgba(255, 255, 255, 0.2)' } },
|
||
splitLine: { lineStyle: { color: 'rgba(255, 255, 255, 0.2)' } }
|
||
},
|
||
series: [{
|
||
name: '技术对比',
|
||
type: 'radar',
|
||
data: [
|
||
{
|
||
value: [95, 40, 60, 30, 20], // 精度高, 灵活性低, 抗干扰一般, 成本高, 场地要求高
|
||
name: '光学动捕',
|
||
areaStyle: {
|
||
color: new echarts.graphic.RadialGradient(0.1, 0.6, 1, [
|
||
{ color: 'rgba(0, 191, 255, 0.5)', offset: 0 },
|
||
{ color: 'rgba(0, 191, 255, 0.1)', offset: 1 }
|
||
])
|
||
}
|
||
},
|
||
{
|
||
value: [70, 90, 40, 60, 95], // 精度较低, 灵活性高, 抗干扰差, 成本中等, 场地要求低
|
||
name: '惯性动捕',
|
||
areaStyle: {
|
||
color: new echarts.graphic.RadialGradient(0.1, 0.6, 1, [
|
||
{ color: 'rgba(138, 43, 226, 0.5)', offset: 0 },
|
||
{ color: 'rgba(138, 43, 226, 0.1)', offset: 1 }
|
||
])
|
||
}
|
||
}
|
||
]
|
||
}]
|
||
};
|
||
techChart.setOption(techOption);
|
||
|
||
// ECharts for Market Share
|
||
var marketChartDom = document.getElementById('market-share-chart');
|
||
var marketChart = echarts.init(marketChartDom, 'dark');
|
||
var marketOption = {
|
||
backgroundColor: 'transparent',
|
||
tooltip: {
|
||
trigger: 'item',
|
||
formatter: '{a} <br/>{b}: {c}%'
|
||
},
|
||
legend: {
|
||
orient: 'vertical',
|
||
left: 10,
|
||
data: ['Vicon (英国)', 'OptiTrack (美国)', '其他'],
|
||
textStyle: { color: '#ccc' }
|
||
},
|
||
series: [{
|
||
name: '高端市场份额',
|
||
type: 'pie',
|
||
radius: ['50%', '70%'],
|
||
avoidLabelOverlap: false,
|
||
label: {
|
||
show: false,
|
||
position: 'center'
|
||
},
|
||
emphasis: {
|
||
label: {
|
||
show: true,
|
||
fontSize: '20',
|
||
fontWeight: 'bold',
|
||
fontFamily: 'Oxanium'
|
||
}
|
||
},
|
||
labelLine: {
|
||
show: false
|
||
},
|
||
data: [
|
||
{ value: 45, name: 'Vicon (英国)' },
|
||
{ value: 40, name: 'OptiTrack (美国)' },
|
||
{ value: 15, name: '其他' }
|
||
],
|
||
color: ['#6366F1', '#3B82F6', '#4F46E5']
|
||
}]
|
||
};
|
||
marketChart.setOption(marketOption);
|
||
|
||
window.addEventListener('resize', function () {
|
||
techChart.resize();
|
||
marketChart.resize();
|
||
});
|
||
});
|
||
</script>
|
||
|
||
</body>
|
||
</html> |