Test the robot when the data stops
A practical ROS 2 fault-injection plan for stale observations, delayed commands and controlled recovery, with an illustrative stopping-distance calculation.
Journal
Notes from the bench.
Practical writing on robotics, embedded systems, AI and software people can trust, plus detailed case studies of my own projects.
23 posts
Inside the Gendio control board: power, protocols and radar positioning
A detailed design study of the ESP32-S3 controller, its power and protocol boundaries, and a radar-based positioning extension that preserves the existing PCB.
Designing a waterless solar-panel cleaning robot
CAD, traction and energy calculations, conservative coverage planning and an experimental roadmap for a modular dry-cleaning crawler.
Building a ROS 2 navigation stack I could investigate
How I organised sensing, estimation, planning and control for a differential-drive rover, and what simulation could establish.
From movement to measurement: my ESP32 honours capstone
Integrating Hall-effect sensing, embedded acquisition, Bluetooth and MATLAB into a movement-assessment prototype.
Making catalogue repricing reviewable at SWL
A local-first pricing workflow with deterministic matching, exact money handling and an operator-reviewed ServiceM8 export.
Designing the Gendio controller around explicit states
An ESP32-S3 display-controller design connecting serial parsing, guarded display output, a four-layer PCB and browser configuration.
The second life of a product
What a product reveals about its design when someone needs to open it, find a part and put it back into service.
In praise of the volume knob
A small control raises a larger design question: how much attention should an everyday adjustment demand?
A smooth sensor trace can hide real motion
How an 80 Hz signal becomes indistinguishable from 20 Hz when sampled at 100 samples per second, and what to check before trusting a trend.
99% accuracy can still miss every defect
A worked inspection example showing why defect recall, false rejects and review workload belong beside the headline accuracy score.
The experience trap: when hiring protects the past
Years in a role can tell us something. They cannot tell us everything. A case for assessing capability, supporting graduates and rewarding people who keep learning.
Before training the model, do the camera maths
A worked example of field of view, exposure and motion blur, and how to tell whether an inspection image contains enough evidence.
The hidden cost of keeping every machine busy
What Little's Law and a simple queue model reveal about work in progress, waiting time and where automation can actually help.
A connected sensor can still be giving you old data
Separating connection health, measurement age and validity in an MQTT dashboard, with a practical stale-data test.
A ROS 2 topic can exist and still deliver nothing
A focused way to diagnose ROS 2 QoS mismatches, separate discovery from delivery, and test the actual subscriber.
Before tuning a robot, check what it thinks is moving
A practical way to separate coordinate-frame, timing and localisation faults before changing a navigation controller.
A noisy sensor needs a decision rule
How hysteresis, debounce and explicit states turn a fluctuating measurement into a useful event without hiding the timing cost.
Local AI: the model is one part of the system
A practical design for checking retrieval, source quality and permission boundaries in an assistant that runs on your own hardware.
The hard part of a CSV import is deciding what may change
Preserving identifiers, resolving ambiguity and making an import reviewable before it changes an operational system.
A local app needs a way back
Why offline use, saved data, export and recovery need separate acceptance tests in a small personal or business app.
When an automation times out, did the action happen?
Designing retries, operation records and human review for the uncertain gap between sending a request and seeing its result.
AI without the jargon: a practical starting point
A plain-English guide to trying AI, protecting your information and checking the answers.