Matic Robot Vacuum Ditches Lidar for 5-Camera Vision, NVIDIA Chip, and Voice Control
Aiden_Tech_Ai · x · 2026-08-14
Matic has introduced its intuitive home robot vacuum, adopting Tesla's pure vision approach by ditching traditional Lidar. Instead, it uses 5 cameras paired with an NVIDIA chip for on-device inference.
- Semantic Recognition: Instead of merely detecting obstacles, it identifies specific items like charging cables, toys, or pet waste, and cleans them accordingly.
- Multimodal Interaction: The new Cues feature supports voice and gesture controls. Users can point at a spill and say "clean this," prompting the robot to locate it in 3D space and handle the task autonomously.
- Background: Backed by $115M in funding and 9 years of R&D, the robot understands 75 languages and is designed to be accessible for everyone from 5-year-olds to 80-year-olds.
Related event: Matic Launches Vision-Only Robot Vacuum, Drops LiDAR for 5 Cameras(14 posts)→
More from Embodied
- Embodied AI Blind Spot: 9 VLMs Know How to Repair but Fail to Proactively Spot Issues — rohanpaul_ai · 2026-08-14
- Meta Donates 15,000 Ray-Ban Meta Smart Glasses to Ireland's Vision Loss Charity — joshto · 2026-08-14
- W&B to Demo World Action Models, Enabling Robots to Adapt to Unseen Tasks — wandb · 2026-08-14
- Tesla Robotaxi Austin Wait Times Drop to 2 Minutes in User Test — elonmusk · 2026-08-14
- EGO1GS: World's First Smart Egocentric Capture Device Announced — YuXiang_IRVL · 2026-08-14
- WARP: Transforming Offline Human Motion into Replayable Whole-Body Robot Actions — chris_j_paxton · 2026-08-14