Robot component layers

This route is a routing tool, not a reference dump. It answers: which layer should I inspect first? Shared-first lanes surface comparison anchors (sensors, connectivity). Signature-heavy lanes preserve branded stacks (AI, voice) for deeper research once the shortlist is tighter. Workbench cards carry preview rows and counts before you open denser directories.

The route is also optimized for research sequencing. Shared layers help establish overlap and comparison anchors. Signature-heavy layers help explain why apparently similar robots still diverge once autonomy stacks or assistant ecosystems enter the picture. The hub exists to restore a reading order so the first ten minutes of research produce a clear next move.

If you are asking "what overlaps?" - start shared-first (sensors, connectivity). If you are asking "what explains the difference?" - start signature-heavy (AI, voice). Shared-first reduces orientation risk; signature-heavy reduces interpretation risk. Start broad, then go deep.

The distinction matters because not every research session starts with the same uncertainty. A buyer comparing robot vacuums may care first about navigation (sensors). An integrator evaluating delivery robots may care first about APIs (connectivity). The lane split gives each session a cleaner first move.

Only 165 of 1735 components repeat across multiple robots. Those carry most early comparison value. One-off entries still matter — but after you know which layer deserves deeper inspection. Collapsing them keeps the reusable signal visible instead of burying it under noise from items that only appear once in the entire database.

Robot count is a browse signal, not a quality score. High counts = comparison anchors (shared building blocks). Low counts = differentiators (proprietary stacks, branded features). Use count to choose reading order, not final judgment.

Pair count with the question you are answering. If you are asking "what is common across the shortlist?" look for higher-count anchors. If you are asking "what makes this manufacturer unusual?" lower-count entries can be the right clue.

AI labels are often vendor-specific and branded — 348 entries, many behaving like autonomy signatures rather than standardized parts. The AI workbench is framed as a signature catalog, not a shared-parts table. Read it like editorial evidence, not a popularity ranking.

That does not make the AI workbench weaker; it makes it different. AI is often where a manufacturer expresses product distinctiveness — one robot may emphasize on-device compute, another whole-home mapping, another OpenAI compatibility. Those labels still matter for research, but they behave less like a reusable table of shared parts and more like a catalog of signature stacks.

A freshness cue: 1230 components tie to recently verified robot records. Use it to triage the most current slice of the catalog — especially useful in fast-moving areas like AI, apps, and assistant integrations where the newest maintained records may answer a question faster than an older untouched slice. Freshness is not the same as quality, but recency helps prioritize attention when you are short on time.

If you are scanning quickly, pair the freshness signal with robot count: repeated components that also show recent activity give you the most current comparison anchors. That combination is faster than reading every label equally.

Yes. Two robots can both list LiDAR, ROS 2, Alexa, or Wi-Fi and still produce very different results. The label tells you a technology appears in the stack — it does not tell you how well it is integrated, how recent the implementation is, or whether the rest of the robot benefits from it.

Use Components to discover the clue, robot pages to inspect the full product context, and Compare when the decision is concrete. Component names are excellent for pattern-finding, but weak on their own as final verdicts. Treat them as evidence that narrows the search space.

Glossary for definitions → Components for pattern-finding → Robot pages for product context → Compare for decisions. If the current page stops changing the decision, leave it.

Each route answers a different class of question. Components tells you whether a technology clue is shared, rare, or current. Robot pages tell you whether that clue sits inside a product that fits your needs. Compare is where those clues get tested side by side against price, battery, and deployment fit.

Ask yourself: am I still narrowing the evidence, or am I already judging products? If you are still narrowing, stay in Components. If you are judging products, move to robot pages or Compare. That distinction keeps research fast and prevents the atlas from becoming a dead-end scroll.

Sensors are about perception quality — navigation reliability, obstacle handling, environmental awareness. If that is the risk, open sensors first. Connectivity is about integration and control — smart-home compatibility, remote operation, deployment friction. AI is about behavior and decision-making — why similar robots act differently once autonomy or branded intelligence enters the picture. Voice affects accessibility, family adoption, and interface comfort.

Match the layer to the risk you are reducing. Worried about navigation quality? Start with sensors. Worried about compatibility? Start with connectivity. Worried about autonomy claims? Start with AI. Worried about household usability? Start with voice. The page is organized around that logic so technical browsing stays tied to real-world decisions.

The fastest way to get value from this route is to match the workbench to the kind of uncertainty you are trying to reduce. Start broad with shared layers (sensors, connectivity) to establish what is actually common across the shortlist. Then go deep with signature layers (AI, voice) to test whether the remaining candidates are still similar once you examine autonomy stacks, assistant ecosystems, or hands-free interaction.

That sequence prevents two common mistakes: reacting to flashy one-off labels too early, and treating the AI layer as a simple popularity contest when it is actually a signature catalog. Start with the anchors, then move to the differentiators.