John Deere's second-gen autonomous tillage: 16 cross-checking cameras beat sensor drift
lukas_m_ziegler · x · 2026-10-15
John Deere sells autonomous tillage as 'an extra pair of hands.' The hard part was never driving straight (AutoTrac solved that) — it was trusting what cameras see on a machine that vibrates, flexes under draft load, and swings tens of degrees in temperature. That trio capped the first-gen system: drifted cameras report wrong depth, so tractors stop for nothing or fail to stop. The second-gen perception system counters with 16 cameras in 360° coverage, deliberate overlap so cameras cross-check and recalibrate in real time, and onboard NVIDIA Orin GPUs for inference.
More from Embodied
- Tester finds GPT-6 Astra still can't solve this robotics problem, predicts it becomes the field's baseline — DJiafei · 2026-10-16
- Dreamscale launches cloud inference for physical AI: 8+ robots per H100, 320ms DreamZero latency — arnie_hacker · 2026-10-16
- WEDGE uses navigation agents' own failures to generate targeted training environments — EliasEskin · 2026-10-16
- Sandbar's Stream Ring named a TIME Best Invention of 2026 in wearables — km · 2026-10-15
- Nvidia Presents Robota-Embed: Unified VLA Embeddings for Robot In-Context Learning — arankomatsuzaki · 2026-10-15
- Chinese startup DexForce trains humanoid robots with virtual data to slash costs — pstAsiatech · 2026-10-15