Interactive tutorials
3 self-contained, interactive guides — 10 pages in all — with live simulations you can poke at: fire a LiDAR pulse, watch a SLAM map drift, corrupt a point cloud, or compare how 30 companies build their data engines. Each opens as a full-screen interactive page.
LiDAR & the Autonomy Stack
~75 min · 6 partsA five-part interactive series walking up the stack from the physical sensor to the data engine everything depends on — point clouds, perception, SLAM, corruption, robustness, and building a simulator. Grounded in published benchmarks; figures current as of early 2026.
- →Series overview
The landing page — read the five parts in order, or jump to whatever you need.
- →Part 1 — How LiDAR works
Light timed to the nanosecond, point clouds, the five hardware types and their parameters, and 2025–26 market pricing. Fire a pulse and watch the round-trip.
- →Part 2 — The autonomy stack
What consumes the points: the perception pipeline, SLAM (build a map live and watch it drift), sensor fusion, and the point-cloud representations that carry meaning into a model.
- →Part 3 — Point clouds as data
How defects become silent model biases. Break a clean cloud six ways in the Corruption Lab — sparsity, occlusion, noise, motion, weather, ghosting.
- →Part 4 — Robustness to sensor failure
The robustness paradox, resolved: corrupt the inputs, never the labels. Drop a sensor and compare a naïve model against a dropout-trained one.
- →Part 5 — Building a LiDAR simulator
Assemble a synthetic cloud stage by stage — raycast, raydrop, intensity, noise — and close the domain gap to a real reference. Why raydrop is the crux.
SLAM & Localization
~45 min · 3 partsThree interactive field guides on how self-driving cars know where they are: a tour of the SLAM back-end (ORB-SLAM, LOAM, Cartographer), the localization spectrum the industry is converging on, and why one LiDAR sweep does localization and perception at once.
- →SLAM Atlas — an interactive field guide
The heart of every modern system — front-end, back-end, filtering vs. smoothing, and the factor graphs (g2o, GTSAM, Ceres) that make it fast.
- →Where am I? — AV localization in 2026
There is no single winner; the frontier is a spectrum, and the industry is quietly converging toward the middle.
- →One sweep, two jobs
Why a single LiDAR scan serves localization and perception together — your three intuitions, graded, and the correction that unlocks the rest.
Physical AI Data Strategy Explorer
~40 minA deep dive into 30 companies across self-driving and robotics in the US, China, and beyond — how they collect real-world data, manufacture synthetic data, auto-label, and evaluate. The 2025–26 shift: the flywheel collapsing into a single foundation/world model that drives, simulates, and evaluates at once.