MHQQysh.github.io/research-bci-multitask-robotics.html at master · MHQQysh/MHQQysh.github.io · GitHub

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
<!DOCTYPE html>
<html lang="zh-CN">
<head>
  <meta charset="UTF-8">
  <meta name="viewport" content="width=device-width, initial-scale=1.0">
  <title>科研项目 | 02 脑机接口引导的多任务机械臂操作 | 袁诗鸿</title>
  <meta name="description" content="02 脑机接口引导的多任务机械臂操作">
  <link rel="icon" href="static/picture/avator.jpg">
  <link rel="preconnect" href="https://fonts.googleapis.com">
  <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
  <link href="https://fonts.googleapis.com/css2?family=DM+Sans:wght@400;500;700&family=Noto+Sans+SC:wght@400;500;700&family=Noto+Serif+SC:wght@400;600;700&family=Playfair+Display:wght@400;600;700&family=Source+Serif+4:opsz,wght@8..60,300;8..60,400;8..60,600&display=swap" rel="stylesheet">
  <link rel="stylesheet" href="style.css">
</head>
<body
  data-page="research"
  data-title-zh="科研项目 | 02 脑机接口引导的多任务机械臂操作 | 袁诗鸿"
  data-title-en="Research Project | 02 BCI-guided Multi-task Robotic Arm Operation | Shihong Yuan"
  data-description-zh="02 脑机接口引导的多任务机械臂操作"
  data-description-en="Research page for 02 BCI-guided Multi-task Robotic Arm Operation."
>
  <header class="site-header">
    <nav class="nav-shell container">
      <a class="brand" href="index.html" aria-label="Shihong Yuan home">
        <span class="brand-mark"><img src="static/picture/avator.jpg" alt="Shihong Yuan avatar"></span>
        <span class="brand-copy">
          <span class="brand-title" data-i18n-en="Shihong Yuan">袁诗鸿</span>
          <span class="brand-subtitle" data-i18n-en="Portfolio archive">作品集档案</span>
        </span>
      </a>
      <button class="nav-toggle" type="button" aria-expanded="false" aria-controls="nav-menu" data-nav-toggle>
        <span></span><span></span><span></span>
      </button>
      <div class="nav-menu" id="nav-menu" data-nav-menu>
        <div class="nav-links">
          <a class="nav-link" href="index.html" data-page-link="home" data-i18n-en="Home">首页</a>
          <a class="nav-link" href="index.html#testi" data-i18n-en="Research">科研</a>
          <a class="nav-link" href="index.html#serv" data-i18n-en="Projects">项目</a>
          <a class="nav-link" href="index.html#blogs" data-i18n-en="Life">生活</a>
          <a class="nav-link" href="index.html#education" data-i18n-en="Education">教育经历</a>
        </div>
        <div class="lang-toggle" aria-label="Language switcher">
          <button class="lang-toggle-btn" type="button" data-lang-option="zh">中文</button>
          <span>/</span>
          <button class="lang-toggle-btn" type="button" data-lang-option="en">EN</button>
        </div>
      </div>
    </nav>
  </header>

  <main>
    <section class="page-hero">
      <div class="project-dense-container page-heading">
        <p class="eyebrow">科研经历</p>
        <h1 class="page-title" data-i18n-en="02 BCI-guided Multi-task Robotic Arm Operation">02 脑机接口引导的多任务机械臂操作</h1>
        <p class="page-summary">
          目标：识别场景为点云，图网络 LLM 推荐执行任务，用户用脑机选择任务，机械臂抓取物体特征点放到另外特征位置执行任务。
        </p>
        <div class="meta-links">
          <span class="pill">2025.06-至今</span>
        </div>
      </div>
    </section>

    <section class="page-shell">
      <div class="project-dense-container story-layout">
        <div class="story-text-column">
          <article class="story-card">
            <h2>01 项目目的</h2>
            <p>目标：识别场景为点云，图网络 LLM 推荐执行任务，用户用脑机选择任务，机械臂抓取物体特征点放到另外特征位置执行任务。</p>
          </article>

          <article class="story-card">
            <h2>02 个人贡献</h2>
            <p>LLM+LightGCN 分类操作任务，实现给一个没见过的物体，也能想到可能对应的任务让用户来选。</p>
            <p>相机标定和 esdf 建图得到物体点云位置的仿真工作。</p>
            <p>EEG 读取脑电信息输出用户的选择的模型训练。</p>
            <p>机械臂执行这部分的仿真工作。</p>
          </article>

          <article class="story-card">
            <h2>03 难点与解决</h2>
            <p>尝试 Semantic Information+LightGCN，最后实现效果超过原算法。</p>
            <p>双相机标定仿真一开始无法重合。后面是发现仿真器中跟现实中相机 z 方向不一样。</p>
            <p>机械臂规划部分的代码，调试了很久。</p>
          </article>
        </div>

        <div class="story-media-column">
          <div class="story-media-card">
            <p class="story-media-title">图片 01</p>
            <img src="static/research/bci/bci-system-architecture.png" alt="BCI project system architecture" loading="lazy">
          </div>

          <div class="story-media-card">
            <p class="story-media-title">图片 02</p>
            <img src="static/research/bci/bci-simulation-scene.png" alt="BCI manipulation simulation scene" loading="lazy">
          </div>
        </div>
      </div>
    </section>
  </main>

  <footer class="site-footer">
    <div class="container footer-grid">
      <div>
        <p class="footer-brand" data-i18n-en="Shihong Yuan">袁诗鸿</p>
        <p class="footer-copy">02 脑机接口引导的多任务机械臂操作</p>
      </div>
      <div class="footer-links">
        <a href="research-gdps-decoupled-posterior-sampling.html" data-i18n-en="Previous project">上一页</a>
        <a href="index.html#testi" data-i18n-en="Research">科研</a>
        <a href="research-force-aware-manipulation.html" data-i18n-en="Next project">下一页</a>
      </div>
      <div class="footer-meta">
        <span>Contact: shihong.23@intl.zju.edu.cn</span>
        <span><span data-i18n-en="Year">年份</span>: <span data-year></span></span>
      </div>
    </div>
  </footer>

  <script src="script.js"></script>
</body>
</html>