Engineering handbook
Robotics Systems Engineering
Practical engineering decisions for designing, selecting, integrating, debugging, testing and deploying robotic systems.
What should I select?Why?How do I integrate it?How do I configure it?How do I verify it works?What happens when it fails?How do I fix it?
The 20 module overviews, the five area hubs and the full syllabus are open. 192 of 192 detailed guides, diagnostics, tools and challenges are written so far; the rest are marked Coming soon and are released phase by phase after testing. ROS 2 commands and parameters are written for Jazzy.
Five ways in
- Engineering GuidesStep-by-step guides for selecting, configuring, integrating and validating robotics hardware and software.98 of 98 open
- Design & SelectionEngineering trade-offs, component selection, hardware sizing, architecture decisions and system design.37 of 37 open
- Troubleshooting & DiagnosticsReal robotics problems with a diagnostic procedure for each: symptoms, measurements, commands, investigation, fix and validation.39 of 39 open
- Interactive Engineering ToolsCalculators, visualizers, simulators, comparison tools and configuration helpers. Preliminary engineering guidance, with stated assumptions.12 of 12 open
- System Design ChallengesComplete engineering problems that need several robotics disciplines at once.6 of 6 open
The 20 modules
Modules are ordered roughly as you would build a robot: requirements, hardware, compute, sensor integration, estimation, autonomy, then deployment. Take any module on its own. Each one links to the FixTheRobot lessons that already teach the theory.
| # | Module | Level | Guides |
|---|---|---|---|
| 01 | Robotics System Architecture & RequirementsTurn an application requirement into a complete robot architecture: requirements, operational design domain, subsystem split, interfaces and an integration plan. | Intermediate | 6/6 |
| 02 | Sensor Selection, Trade-offs & Integration FlagshipChoose, place, wire, drive, calibrate and validate sensors. Each sensor is covered by one fixed template, from what it measures to how it fails, with product facts taken only from manufacturer datasheets. | Intermediate | 22/22 |
| 03 | Actuators, Motors, Electrical & Power IntegrationSize motors and gearboxes, budget voltage and current, choose a battery, estimate runtime and keep electrical noise out of sensor measurements. | Intermediate | 8/8 |
| 04 | Embedded Systems & Hardware InterfacesChoose and debug the link between a sensor or actuator and the computer: UART, SPI, I2C, CAN, CAN-FD, Ethernet, USB. Includes wiring, configuration and reliable driver design. | Intermediate | 7/7 |
| 05 | Onboard Computer, CPU/GPU Selection & Compute SizingEstimate sensor bandwidth, memory and processing load, choose x86, ARM or Jetson, and find the bottleneck with standard Linux and ROS 2 tools. | Intermediate | 5/5 |
| 06 | Camera–LiDAR Calibration & Coordinate Transformations FlagshipGet frames and extrinsics right: PL = TLC × PC with the direction of every transform spelled out, camera projection, calibration validation and TF debugging. | Intermediate | 7/7 |
| 07 | Sensor Time Synchronization & Latency EngineeringSeparate clock synchronization, measurement synchronization, transport latency and processing latency, then measure and fix each. | Advanced | 6/6 |
| 08 | Robotics Perception & Sensor IntegrationBuild and debug an obstacle-detection pipeline from camera and LiDAR: filtering, ground segmentation, clustering, detection, association, tracking and throughput. | Intermediate | 6/6 |
| 09 | SLAM Selection, Configuration & TroubleshootingPick a SLAM system for the sensors and compute you have, configure it, and diagnose drift and tracking loss. ORB-SLAM3, VINS-Fusion, OpenVINS, LIO-SAM, FAST-LIO2 and KISS-ICP are compared from their official documentation and repositories. | Advanced | 5/5 |
| 10 | Visual-Inertial SLAM & LiDAR-Inertial IntegrationThe hands-on side of VIO and LIO: IMU noise parameters, camera–IMU and LiDAR–IMU extrinsics, time offsets, initialization and recovery. | Advanced | 6/6 |
| 11 | Sensor Fusion & State EstimationConfigure EKF-based localization with robot_localization, tune covariances, reject bad measurements and decide between filters and factor graphs. Not every SLAM system is a Kalman filter, and the guides say which are not. | Advanced | 5/5 |
| 12 | Outdoor-to-Indoor Localization & GNSS-Denied Navigation FlagshipFlagship module. A robot leaves GNSS coverage and enters a building with no pose jump. This is a covariance-aware estimator handover with aligned frames and a recovery plan, not a SLAM on/off switch. | Advanced | 6/6 |
| 13 | Navigation Stack & Autonomous Motion PlanningSelect planners and controllers, configure Nav2 on ROS 2 Jazzy, tune costmaps and footprints, set up recovery behaviors and validate navigation in a real building. | Advanced | 6/6 |
| 14 | Motion Control & Actuator IntegrationTune and validate controllers on a real robot: PID, cascaded loops, odometry calibration, wheel slip, saturation and ros2_control. | Intermediate | 9/9 |
| 15 | Simulation-to-Real IntegrationMake a robot model and simulated sensors match the hardware, and find out why something that works in simulation fails on the robot. | Intermediate | 5/5 |
| 16 | Manipulation & Robotic Arm IntegrationIntegrate a camera, gripper and arm: hand–eye calibration, object pose to grasp pose, MoveIt 2 setup and diagnosing failed picks. | Advanced | 5/5 |
| 17 | System Performance & OptimizationFind where latency, CPU, GPU, memory and bandwidth go, and fix message loss with ROS 2 QoS, intra-process communication and executors. | Advanced | 5/5 |
| 18 | Robotics Safety, Reliability & Failure HandlingDetect failures, reach a safe state and prove it. Separates ordinary operational safeguards from safety-rated functions; software alone does not make a robot functionally safe. | Advanced | 5/5 |
| 19 | Testing, Validation & Field DeploymentCommission and field-test a robot: pre-power checks, sensor and calibration validation, bag recording, reproducing intermittent failures, regression tests, deployment and rollback. | Intermediate | 6/6 |
| 20 | Fleet Management & Operational RoboticsRun many robots: task allocation, traffic management, shared maps, remote health monitoring, OTA updates and long-term maintenance. | Advanced | 5/5 |
How to use this handbook
- Designing a robot? Start at Module 01, then use the Design & Selection guides for sensors (02), power (03) and compute (05).
- Something is broken? Go to Troubleshooting and search by symptom. Every entry is a diagnostic procedure with measurements and commands, not a list of guesses.
- Need a number? The tools state their assumptions and are labelled preliminary engineering guidance. They never replace a datasheet.
- Want theory first? Each module links the existing SLAM, Perception and Navigation lessons instead of repeating them.
What to trust here
- Product facts (sensor and board specifications, software features) come only from official datasheets, documentation and repositories, cited with the date they were checked. Where a source could not be checked, the guide says which datasheet field to look up instead of stating a number.
- General engineering guidance is labelled as such and states its assumptions. Formulas shown on a page are the ones the tools use and are tested.
- Safety: software safeguards on these pages are ordinary operational measures. They are not safety-rated functions and do not make a robot functionally safe on their own.