PRIVACY, LIDAR, AND “UNDER THE RIM”: HOW I CHOSE A ROBOT TO CLEAN A PUBLIC RESTROOM (AND ALMOST RECONSIDERED MY LIFE CHOICES)

It didn’t start with a budget.
It didn’t start with “digital transformation.”

It started when I once again heard the phrase:
“There’s an issue in the restroom again… you know what I mean.”

And that’s when something clicked.

You can operate the best facility in the city — flawless architecture, premium service, curated lobby music, polished brand strategy — but if the restroom feels neglected, none of it matters. What stays in people’s minds is painfully simple: “ugh.” Psychology is merciless.

That’s how I found myself deep in the market of public restroom cleaning robots.
And yes, it sounds like a joke. It isn’t.

Why I Even Looked at Robots (Instead of “Just Hiring More Staff”)

Three reasons. None of them romantic.

1) People are running out.

Cleaning labor is in short supply, and restroom cleaning is its own circle of… not even hell — more like chronic workforce scarcity. Shift stability is a lottery.

2) The restroom is a reputation touchpoint.

It’s not “back-of-house.” It’s a brand mirror.
Especially in malls, transport hubs, universities, hospitals. One bad day — and complaints start flowing. Reviews. Stories. Sometimes with a geolocation tag. Thank you very much.

3) I needed predictability and accountability.

“We cleaned it” is not a metric.

A metric is:

• cycle time
• downtime
• repeat interventionsXZ
• incidents
• complaints before/after

And this is where robots stop sounding like toys and start looking like quality-control infrastructure.

Market Reality: It’s Still Green — and That Matters

The biggest mistake? Confusing floor-cleaning robots with restroom-cleaning robots.

Floor robots are mature. Scaled. Everywhere.

Robots that actually enter stalls, clean toilets and urinals, open doors, manage water systems — that’s still an early-stage market. Pilots. First commercial deployments. And plenty of “it will be perfect, we promise.”

(Sometimes it is. Sometimes… well.)

And that’s normal. A restroom is one of the most difficult physical environments for robotics:

Tight. Wet. Human traffic. Non-standard geometry. Edge cases that make you ask, “Why is this even designed like that?!”

1) The Market Map — All the Players, Categorized

Otherwise, you drown in noise.

Category A — Whole-Restroom / Mobile Manipulators

“Robot cleans the entire restroom.”
Wheeled base + robotic arm + tools + docking + water + reporting.

1. Primech AI

• Country: Singapore (SG)
• Product: HYTRON
• Revenue Model: RaaS / demo-based engagements
• Publicly Claimed Capabilities: Toilet, urinal, sink, mirror and floor cleaning; door interaction; refill/discharge systems; self-charging
• Notes: North America expansion announced toward CES 2026

2. Somatic

• Country: United States (US)
• Product: (Model name not always publicly specified)
• Revenue Model: $1,500 / $3,000 / $4,500 per month subscription
• Publicly Claimed Capabilities: Water + chemical application, wet-vac functionality, door handling, autonomous operation with VR-based setup
• Notes: Transparent pricing publicly available on website

3. HiveBotics

• Country: Singapore (SG)
• Product: Abluo
• Revenue Model: Not disclosed
• Publicly Claimed Capabilities: 3D navigation, robotic manipulator, steam/brush systems, focus on high-contamination (“dirty”) zones
• Notes: Pilots and commercial launches discussed for 2024–2025

4. Loki Robotics

• Country: US / Switzerland (US/CH)
• Product: Loki
• Revenue Model: Pilot-based deployments
• Publicly Claimed Capabilities: Tool changing, “cleaning like a person,” teleoperation combined with machine learning
• Notes: Hybrid autonomy philosophy

5. Cleaning Robotics

• Country: Not clearly disclosed
• Product: TCR series
• Revenue Model: Early access / demo programs
• Publicly Claimed Capabilities: “Visual AI” positioning for restroom cleaning
• Notes: Legal entity transparency varies

6. Peanut Robotics (adjacent)

• Country: United States (US)
• Product: General-purpose robotic system
• Revenue Model: Commercial sales claims
• Publicly Claimed Capabilities: Manipulator-based platform claiming to clean “everything”
• Notes: Not restroom-specific, but competes within the same facilities budget discussion

Category B — Privacy-First / Linear High-Volume Cleaning

Designed for large public restrooms where privacy is non-negotiable.

1. ZeeqClean

• Country: Hong Kong (HK)
• Product: ZC-01
• Core Proposition: Non-vision navigation + contactless jet cleaning + drying + UV disinfection
• Why It Matters: Cameras in restrooms are a reputational landmine

2. HKPC

• Country: Hong Kong (HK)
• Product: Smart Public Toilet Cleaning Robot
• Core Proposition: IoT + SLAM + vision-based navigation
• Why It Matters: Public sector deployments announced

Cameras in restrooms — even “technical, non-recording ones” — require explanations.
And people don’t like explanations in bathrooms.

ZeeqClean’s privacy-by-design positioning is strategically smart.

I literally wrote in my notes:

“Let the robot see less than we end up seeing in headlines.”

Harsh. But realistic.

Category C — In-Bowl / Attach-On Devices

Increase cleaning frequency. Rarely replace staff.

1. Altan Robotech

• Country: United States (US)
• Product: Giddel
• Pricing / Model: $449.99 (listed)
• Nuance: Works well at small scale; deploying across 30 stalls becomes operationally complex

2. Toibot

• Country: Israel (IL)
• Product: TOIBOT
• Pricing / Model: Historically positioned as low-cost (recent pricing not publicly verified)
• Nuance: Registry documentation mentions integrated disinfecting tablet mechanism

3. SpinX

• Country: Israel (IL)
• Product: SPINX
• Pricing / Model: Kickstarter project
• Nuance: Promised 90-second cleaning cycle; did not scale to mass market

These are pragmatic tools — not workforce replacements.

Category D — Teleoperated / Avatar Robots

When autonomy isn’t reliable enough — add a remote operator.

Mira Robotics / ugo
• Country: Japan (JP)
• Product: ugo
• Known Facts: Reported ~$1,000 per month lease; one operator can control up to four robots

Hybrid control can reduce operational risk in unpredictable environments.

And restrooms are unpredictable.

Category E — Not Restroom-Specific, But in the Same Budget Fight

Because the CFO will ask:

“Why not just buy another floor robot?”

1. Gausium

• Country: China / Global (CN/Global)
• Product: Phantas (and other floor-cleaning models)
• Why Relevant: Large-scale floor robot deployments that cover restroom perimeters and compete for the same facilities automation budget

2. Zerith Robotics / Zerith AI

• Country: China (CN)
• Product: H1
• Why Relevant: Wheeled humanoid for housekeeping, including restroom cleaning; media references mention deployments in “20+ locations”

3. Jingwu Robotics

• Country: China (CN)
• Product: 3D Cleaning Robot
• Why Relevant: 360° restroom cleaning demonstrations; early-stage positioning but relevant as competitive signal/noise

2) How I Evaluated the Market — By Risk, Not Branding

You don’t buy a restroom robot for “AI.”

You buy it to reduce chaos.

Here’s what truly matters.

Criterion #1 — Privacy

High-sensitivity environment.

If cameras are involved — even “for navigation only” — prepare to justify it to users, management, possibly regulators.

Privacy-by-design reduces exposure.

Criterion #2 — Infrastructure: Water, Drainage, Docking, Consumables

This is where 70% of beautiful presentations collapse.

• Where is the dock installed?
• Is water access available?
• Where does wastewater go?
• Who refills consumables? How often?
• What happens if the dock is blocked by a janitor’s cart? (Spoiler: conflict.)

Refill/discharge systems and self-charging are not “features.”

They are survival conditions.

Criterion #3 — Safety in Wet Zones

Uncomfortable thought:

A robot can clean perfectly… and create the perfect slipping hazard.

Two critical questions:

• How is water usage controlled?
• What happens if a person enters mid-cycle? Stop? Reverse? Continue like a tank?

In public spaces, safety tolerance is zero.

Criterion #4 — Full Autonomy vs Hybrid Reliability

Two camps:

“We are fully autonomous.” (Sounds great.)
“We are hybrid: autonomy + telepresence/learning.” (Sometimes more honest.)

Loki openly promotes hybrid learning in high-variation environments.
ugo follows a different philosophy entirely: operator-managed fleets.

My conclusion?

In restrooms, “perfect autonomy” often loses to “predictable hybrid.”

Not futuristic. But operationally stable.

3) Comparative View — Who Covers What

Not “who is best,” but “who fits what.”

HYTRON

• Full restroom cleaning: Yes (claimed)
• Privacy without cameras: Not clearly public
• Revenue model: RaaS
• Stage: Active demos

Somatic

• Full restroom cleaning: Yes
• Privacy without cameras: Not clearly public
• Revenue model: Fixed subscription
• Stage: Commercial

Abluo (HiveBotics)

• Full restroom cleaning: Yes (pilots)
• Privacy without cameras: Not clear
• Revenue model: Not disclosed
• Stage: Early commercial

ZC-01 (ZeeqClean)

• Full restroom cleaning: Focus on stalls/urinals + basic floor
• Privacy without cameras: Yes (non-vision navigation)
• Revenue model: Quote-based
• Stage: Early commercial

Loki

• Full restroom cleaning: Target
• Privacy without cameras: Not clear
• Revenue model: Pilot-based
• Stage: Early

ugo (Mira Robotics)

• Full restroom cleaning: Operator-dependent
• Privacy without cameras: Configuration-dependent
• Revenue model: Lease / service
• Stage: Historical commercial model

In-bowl Devices (Giddel / Toibot / SpinX)

• Full restroom cleaning: Bowl only
• Privacy without cameras: Yes
• Revenue model: Device purchase
• Stage: Consumer / niche

4) My Practical Selection Logic

No universal winner.

It depends on the facility.

High-traffic malls / airports / transit hubs

Look at:

• Privacy-first solutions
• Or whole-restroom robots — but only after infrastructure audit

Offices / casinos / campuses

Somatic’s transparent pricing is rare and operationally convenient.

Complex layouts / unpredictable usage

Hybrid logic (Loki / ugo philosophy) may outperform “pure autonomy.”

Small business, 1–4 stalls

In-bowl solutions like Giddel are pragmatic.

But manage expectations.
It’s not “robot instead of a team.”
It’s “robot instead of part of the dirty routine.”

5) Pilot KPIs — What I’d Put in the Contract (Strictly, but Fairly)

Without this, a pilot becomes a showroom.

• Cycle time per stall / restroom
• % successful cycles without human intervention
• Downtime and root causes
• Safety incidents
• Water/chemical consumption per cycle
• Complaints before/after
• Mean time to recovery (SLA compliance)

And yes — one uncomfortable KPI:

“How does the robot behave when a stall contains something we’d rather not describe?”

Because that’s exactly what will happen.

A Small but Honest Reflection

At some point during this research, I caught myself thinking:

We market robots as revolutions.
But we buy them to stop firefighting.

And another thought — simpler:

In a restroom, you don’t need the smartest robot.

You need the most predictable one.

That’s the real difference.