From the mechanics of motion to the neural networks that give machines perception — a complete look at the technology underneath modern robotics.
Electric motors, hydraulic cylinders, pneumatic systems — every robot chooses how to generate force. The choice shapes everything: precision, speed, power density, and what the robot can actually do in the world.
A camera is the easy part. The hard part is everything that happens next — turning a flood of raw sensor data into something a machine can use to understand where it is, what surrounds it, and what to do about it.
Structured light, time-of-flight, and stereo vision — coming March 2025.
For decades, industrial robots were choreographed — every movement scripted in advance. Large-scale neural models trained on diverse manipulation data are changing the fundamental paradigm of how robots acquire new skills.
The challenge of getting a robot from A to B without hitting anything is deceptively complex. We map the major algorithmic families — grid-based, sampling-based, and optimisation-based — and explain what makes each one suited to different environments.
Going beyond planning to find the best path — coming April 2025.
Traditional robots are rigid — precise, powerful, but dangerous near humans. A growing research field is building machines from flexible, deformable materials that can adapt to their environment in ways no rigid system can match.
Brain-inspired chips for robotic control — coming Q2 2025.