Solid-State 3D Perception Gateway | Yongkui Xu (MRDVS)
Yongkui Xu, identified in the talk as CEO of Iris, presents Humanoids Summit on why robots need 3D perception and describes Iris’s solid-state LiDAR and robot vision products.
In this talk
- 0:37
Why humanoids need better perception
Xu says humanoids must understand unstructured, changing environments, unlike AGVs and AMRs that can follow fixed paths. He argues that the gap between walking and understanding the world limits task execution, even for activities such as playing football.
- 4:33
Cameras and solid-state LiDAR
Xu contrasts camera-based spatial perception with LiDAR, which directly measures distance. He says rotating LiDAR can be bulky, vulnerable to shock, and limited in service life, while solid-state LiDAR has no rotating parts, is compact, and can combine distance with RGB texture.
- 6:07
Iris’s robot vision capabilities
Xu describes Iris, which he expands as Mobile Robot Vision System, as a company focused on 3D vision for robots including lawn mowers and commercial cleaning robots. He says its capabilities span optics, calibration, addressing, and AI algorithms, with applications in mobile, humanoid, and industrial robots.
- 8:02
Products and humanoid sensor placement
Xu gives product examples: the M4 provides millimeter accuracy at one meter, and the S11 measures to five meters with accuracy within 10 millimeters. For humanoids, he assigns environmental awareness to head sensors, close-range grasping and manipulation at 0.1 to 1 meter to wrist sensors, and a backup role to waist or chest sensors.
Briefs summarise talks and interviews; timestamps open the video at that moment. Claims in a talk are the speaker's own.