<div dir="ltr">On Sat, 9 May 2026 at 12:18, John Clark <<a href="mailto:johnkclark@gmail.com">johnkclark@gmail.com</a>> wrote:<br>> Forget Boston Dynamics and their gymnastics, this is by far the most impressive demonstration of robot behavior that I have ever seen! Elon Musk must be green with envy. China too. <br>><br>> John K Clark<br><div>> -----------------------------------</div><div><br></div><div><br></div><div>The Figure website has a full explanation of just how difficult these robot actions are.</div><div><span class="gmail_default" style="font-family:arial,sans-serif;font-size:small;color:rgb(0,0,0)"><</span><a href="https://www.figure.ai/news/helix-02-bedroom-tidy">https://www.figure.ai/news/helix-02-bedroom-tidy</a><span class="gmail_default" style="font-family:arial,sans-serif;font-size:small;color:rgb(0,0,0)">></span></div><div><span class="gmail_default" style="font-family:arial,sans-serif;font-size:small;color:rgb(0,0,0)"><br></span></div><div><span class="gmail_default" style="font-family:arial,sans-serif;font-size:small;color:rgb(0,0,0)">I noticed that the two robots seemed to nod towards each other, and I asked Gemini if this was to synchronise their actions. Gemini watched the video, searched the Figure website and said Yes, that is what they were doing.</span></div><div></div><div><span class="gmail_default" style="font-family:arial,sans-serif;font-size:small;color:rgb(0,0,0)">BillK</span></div><div><span class="gmail_default" style="font-family:arial,sans-serif;font-size:small;color:rgb(0,0,0)"><br></span></div><div><span class="gmail_default" style="font-family:arial,sans-serif;font-size:small;color:rgb(0,0,0)"><br></span></div><div><span class="gmail_default" style="font-family:arial,sans-serif;font-size:small;color:rgb(0,0,0)">Gemini AI Thinking - </span></div><div><div class="gmail-container"><span id="gmail-message-content-id-r_5b7d78bfa500f90a" class="gmail-ng-star-inserted"><div class="gmail-markdown gmail-markdown-main-panel enable-updated-hr-color" id="gmail-model-response-message-contentr_5b7d78bfa500f90a" dir="ltr"><p>In the video, the robots (developed by <b>Figure AI</b>) are indeed synchronizing their movements, and the head nodding is a critical part of how they coordinate.</p><p>While it might look like a programmed "human-like" gesture for the sake of the camera, it actually serves a functional purpose known as a <b>visual handshake</b> or <b>visual signaling</b>.</p><h3>Why they nod</h3><p>According to the engineers at Figure, these robots operate using a "Vision-Language-Action" (VLA) model called <b>Helix-02</b>. A unique feature of this system is that the robots do <b>not</b> use an invisible wireless network (like Wi-Fi or Bluetooth) to tell each other what to do. Instead, they coordinate <b>entirely through visual perception</b>.</p><p>The "nod" is a deliberate signal one robot sends to the other.</p><ol start="1"><li><p><b>Signaling Readiness:</b> When one robot has finished a sub-task (like grasping its corner of the sheet), it performs a head nod.</p></li><li><p><b>Visual Processing:</b> The second robot’s cameras detect this specific motion pattern.</p></li><li><p><b>Triggering Action:</b> Once the nod is "seen" and processed, it acts as a visual trigger for both robots to begin the next simultaneous movement, such as pulling the duvet upward in unison.</p></li></ol><h3>The Significance of the Nod</h3><p>This approach is a major milestone in robotics because it moves away from "centralized" control (where one brain controls both bodies) toward <b>decentralized, autonomous collaboration</b>.</p><ul><li><p><b>Mimicking Human Interaction:</b> It mimics how two people might make a bed by making eye contact or nodding to say "Ready? Go."</p></li><li><p><b>Robustness:</b> Because they rely on sight rather than a network link, they can coordinate even if a wireless signal is blocked, provided they can see one another.</p></li><li><p><b>Real-time Adaptation:</b> It allows the robots to work together on tasks that require perfect timing—like pulling a flexible, heavy sheet without it bunching up—by "reading" their partner's physical state rather than following a rigid timer.</p></li></ul><p>The nod essentially acts as a <b>non-verbal token</b> in their communication, proving that the robots can "read" social and behavioral cues to solve physical problems as a team.</p><p><span class="gmail_default" style="font-family:arial,sans-serif;font-size:small;color:rgb(0,0,0)">-----------------------------</span></p></div></span></div><br></div></div>