This blog is my robotics field guide: notes and tools from building robots — what worked, what broke, and what I learned.
It’s not a “perfect tutorial” blog. It’s a lab notebook you can actually use.
Why this exists
Robotics is a stack of stacks: software, physics, sensors, math, electronics, debugging, and the occasional existential crisis when Gazebo launches upside-down.
So I’m building a system that helps me (and maybe you) learn like a builder:
- Capture what happened (even the mistakes)
- Extract patterns and checklists
- Reuse them in the next project
- Improve the mental model each loop
What you’ll find here
- Build logs (real progress, real mistakes, real fixes)
- Toolbox posts (reusable patterns, templates, checklists)
- Clear explanations that don’t rely on memorization
The learning principles behind this blog
I don’t learn well by “memorize first, understand later.” For me, learning sticks when it’s built on three principles:
1. Experience - (build memories you can reuse)
Episodic memory bias: You remember what you lived (or vividly imagined living) more than what you only “studied.”
Mental simulation strength: The “MIND strengths” often work like a scene-building engine. You take fragments of experiences and recombine them to model how systems work, predict outcomes, and solve problems.
Material Reasoning: “What object/space picture is this?”
Interconnected Reasoning: “What does this connect to that I already know?”
Narrative Reasoning: “What’s the story of failure → fix?”
Dynamic Reasoning: “How does it change over time if I tweak a parameter?”
2. Understanding - (connect it to the big picture)
Facts don’t stick in isolation. If I can’t connect an idea to why it exists, where it fits, and what it affects, it evaporates. My way of learning is meaning-first: I build understanding by building a network, not by collecting definitions. (That’s basically “meaningful learning” + schema building.)
So whenever I learn something in robotics, I force three anchors:
System architecture: where it lives in the stack (nodes, TF, costmaps, sensors)
Underlying principle: the rule underneath (control, estimation, geometry, signals)
Practical consequence: what breaks if I ignore it (failure modes + debugging clues)
I also start with a short roadmap before the details (big picture → then zoom in). That preview gives my brain a place to attach the new information—like adding the map before exploring the territory.
3. Prediction - (learning by testing models)
My strongest learning happens through prediction errors: I form a guess about what will happen next, test it, and let the mismatch (the “surprise”) teach me. The brain is basically a prediction machine that updates its internal model when reality disagrees.
When my prediction is correct, I move on. When it’s wrong, my attention spikes and the lesson sticks—because prediction error is tightly linked to the brain’s learning/reward machinery (including dopamine-based error signals).
So instead of swallowing facts like pills, I learn best when I’m shown a case, asked to infer the rule, then asked to predict what happens in a new situation—and debug my thinking when I’m wrong.
That’s why I use this loop in almost every project:
The IDEA Loop
I use a simple loop to make this practical:
Imagine — simulate possible outcomes using what I already know
Decide — pick the most likely outcome (my prediction)
Explore — test it in the real world (or a sim / example)
Assess — compare result vs prediction; if wrong, find why and update the model, then repeat
What makes this “Field Guide” style
When I explain something, I try to do three things:
- Start big-picture: what problem are we solving and why?
- Zoom into atoms: what each part does (code, params, frames, signals)
- End with reuse: a checklist, template, or pattern you can copy
If a post includes code, it usually includes:
- What it does
- Why it’s structured that way
- Common failure modes & troubleshooting
Who this is for
- People building with ROS 2, Nav2, Gazebo/Isaac, sensors, and real robots
- People who want reasoning-first learning
- People who like honest notes more than polished marketing
Where to start
If you’re new here:
- Start with the Build Logs tag for the story of what I’m building
- Use Toolbox posts when you want fast reusable patterns
We are not here to become repositories of facts.
We are here to become explorers — mental simulators who can see the terrain ahead, connect distant ideas, and chart paths into the unknown.Details matter.
But they serve the vision.
Welcome to the Field Guide…
"Do. Fail. Learn. The rest will follow. — Tony Fadell"