Robotics & Perception · An independent companion

The Robotics Laboratory

Turn the book’s numerical examples into experiments. Adjust a measurement, move a robot, or step through an algorithm—and inspect what changes.

33Section laboratories
6Robot chapters
OfflineEvery page is self-contained

2. The trash sorter

SECTION 2.1

Modeling the World State

PMFs, cumulative distributions, and sampling

SECTION 2.2

Actions for Sorting Trash

Cost distributions and expected action costs

SECTION 2.3

Sensors for Sorting Trash

Conditional densities and three-sensor simulation

SECTION 2.4

The Perception Engine

Likelihoods, posterior curves, and adjustable MAP sensor fusion

SECTION 2.5

Decision Theory

Compare naive, minimax, and Bayes decisions

SECTION 2.6

Learning from Data

Counts, smoothing, conditional models, and Gaussian fitting

3. The robot vacuum

SECTION 3.1

The Vacuum’s State

A probability distribution over the five rooms

SECTION 3.2

Actions over Time

Controlled Markov chains and battery dynamics

SECTION 3.3

Dynamic Bayesian Networks

Light sensing and ancestral simulation

SECTION 3.4

Perception with Graphical Models

MAP trajectories and smoothed HMM marginals

SECTION 3.5

Markov Decision Processes

Rewards, control tapes, and exact policy values

SECTION 3.6

Learning to Act Optimally

Value iteration, policy iteration, and Q-learning

4. Warehouse robots

SECTION 4.1

Continuous State

Gaussian densities, covariance, mixtures, and samples

SECTION 4.2

Moving in 2D

Omni-wheel kinematics and uncertain motion

SECTION 4.3

Continuous Sensor Models

Warehouse proximity, RFID range, and GPS likelihoods

SECTION 4.4

Localization

Bayes filtering, particles, least squares, and Kalman smoothing

SECTION 4.5

Planning for Logistics

Value iteration on the notebook’s warehouse map

SECTION 4.6

System Identification

Scalar and multivariate least-squares estimation

5. The differential-drive robot

SECTION 5.1

Differential-drive State

Pose, orientation, and uncertainty in SE(2)

SECTION 5.2

Differential-drive Motion

Forward and inverse wheel kinematics

SECTION 5.3

Cameras for Robot Vision

Pinhole projection, calibration, and field of view

SECTION 5.4

Computer Vision 101

Convolution, gradients, and recorded neural vision results

SECTION 5.5

Path Planning

Grow and inspect a rapidly exploring random tree

SECTION 5.6

Deep Learning

Interpolation and training a learnable line grid

6. Autonomous vehicles

SECTION 6.1

Planar Geometry

Rotations, rigid transforms, and noncommutative composition

SECTION 6.3

Sensing for Autonomous Vehicles

Ray intersections and real LIDAR scans

SECTION 6.4

SLAM

Pose-graph optimization and landmark constraints

7. Autonomous drones

SECTION 7.1

Moving in Three Dimensions

3D transforms, rotation order, and angle conventions

SECTION 7.2

Multi-rotor Aircraft

Tilt, kinematics, thrust, and flight dynamics

SECTION 7.3

Sensing for Drones

Camera frames, image projection, and stereo depth

SECTION 7.4

Visual SLAM

IMU integration, bias drift, and recorded reconstruction

SECTION 7.5

Trajectory Optimization

Obstacle costs, path optimization, and thrust allocation

SECTION 7.6

Neural Radiance Fields

Ray sampling, volume rendering, interpolation, and recorded NeRF results

About this collection

Each laboratory starts from the notebook’s numerical models and includes the relevant Python code. Randomized experiments expose a reproducible seed. Recorded outputs from large pretrained models are labeled as recorded results; they are not live neural-network inference.

The collection covers every section containing substantive executable Python. Introductions, summaries, and sections containing only imports or comments are omitted. You can save or copy an individual HTML file and use it without installing Python or connecting to a server.