
One of the strongest current RightRobot candidates for this use case: RTINGS explicitly reports excellent pet-waste avoidance, and Dreame's US Complete package contains the tested X50 Ultra robot body.
Open robot profile →If a dog or cat sometimes has accidents indoors, pet-waste avoidance is not a minor smart-home feature. A single miss can spread solid waste through wheels, brushes and across the floor. RightRobot therefore treats this differently from generic obstacle avoidance: for a strict must-have, we require strong exact-model independent evidence rather than trusting an AI or camera claim on the specification sheet.

Look for explicit pet-waste testing, not just a generic obstacle-avoidance score or a camera/AI badge.
RightRobot's strict matcher fails closed: manufacturer claims alone do not satisfy an 'avoid pet accidents' must-have.
Floor type matters because some obstacle systems perform much worse on carpet than hard flooring.
Solid-waste avoidance does not prove safe handling of urine, vomit, diarrhoea or other wet messes; check the floor before an unattended run whenever an accident is plausible.
This evidence is about recognising/avoiding solid simulated pet waste. Urine, vomit, diarrhoea and other wet messes are different hazards and should not be treated as covered by a solid-waste avoidance result.
RightRobot has already analysed these tests as review evidence. A pass applies to that particular setup; it is not a guarantee that every real pet accident will be avoided.

Avoided fake poo, socks and power cords in both lit and dark test conditions. RightRobot separately re-verified this test segment.
Watch the test at 05:31 ▶
Recorded a 100% / zero-touch obstacle result including artificial pet waste, shoes, mugs and a power cord.
Watch the test at 03:06 ▶
Avoided pet waste and several other obstacles, but the reviewer also observed inconsistent close approaches/touches in the wider obstacle test.
Watch the test at 05:08 ▶
Bumped smaller objects including the simulated pet waste rather than avoiding them.
Watch the test at 04:05 ▶
▶Poo test: passDreame X40 UltraJamie Andrews · 05:51Watch test ↗
▶Poo test: passiRobot Roomba j7+Cordless Vacuum Guide · 01:25Watch test ↗
▶Poo test: passMOVA P20 UltraJust A Dad Approved · 05:01Watch test ↗
▶Poo test: passNarwal Flow 2Jamie Andrews · 03:38Watch test ↗
▶Poo test: passNarwal Flow 2Just A Dad Approved · 07:25Watch test ↗
▶Poo test: passRoborock S8 MaxV UltraModern Castle · 03:25Watch test ↗
▶Poo test: passRoborock S8 MaxV UltraJamie Andrews · 07:25Watch test ↗
▶Poo test: failed / contactedECOVACS DEEBOT X2 OmniModern Castle · 05:46Watch test ↗
▶Poo test: failed / contactedRoborock Q8 Max+Jamie Andrews · 05:20Watch test ↗RightRobot links to the original reviewer video rather than copying or re-hosting the footage.
These are starting points, not a universal ranking. The personalised matcher can promote a different robot when your home gives different requirements more weight.Starting points only — Home Fit can change the order for your home.

One of the strongest current RightRobot candidates for this use case: RTINGS explicitly reports excellent pet-waste avoidance, and Dreame's US Complete package contains the tested X50 Ultra robot body.
Open robot profile →
RTINGS reports highly consistent pet-waste recognition and avoidance on both bare floor and carpet, giving it strong exact-model evidence for this specific risk.
Open robot profile →
RTINGS' exact E28 test avoided pet waste, although it came very close; RightRobot keeps that caveat visible rather than turning a pass into a guarantee.
Open robot profile →
An important benchmark: Vacuum Wars reported a perfect simulated pet-waste test, while iRobot backs the j7/j7+ with a limited Pet Owner Official Promise for solid cat or dog waste.
Open robot profile →20 models checked against exact-model pet-waste evidence.
For a strict pet-waste must-have, RightRobot requires strong exact-model independent evidence and fails closed when that evidence is missing or mixed.
Manufacturer documentation may come from another regional site when it describes the same robot body. RightRobot treats that as supporting product evidence, not as evidence of US availability or price.
Strong exact-robot independent evidence: RTINGS reports that the X50 Ultra's camera + 3D structured-light system detects and avoids pet waste very reliably.
+Strong exact-robot independent evidence: RTINGS reports that the X50 Ultra's camera + 3D structured-light system detects and avoids pet waste very reliably.
The independent test is for the X50 Ultra robot body; package contents can vary by market; no robot should be treated as infallible around animal waste.
Eligible for strict must-haveStrong exact-model independent evidence: RTINGS describes the S1 Pro as highly consistent at identifying and avoiding pet waste on both bare floor and carpet.
+Strong exact-model independent evidence: RTINGS describes the S1 Pro as highly consistent at identifying and avoiding pet waste on both bare floor and carpet.
Strong test evidence reduces risk; it does not make unattended pet-accident avoidance a guarantee.
Eligible for strict must-haveStrong independent pet-waste test evidence plus iRobot's unusual P.O.O.P.
+Strong independent pet-waste test evidence plus iRobot's unusual P.O.O.P. replacement promise for solid cat or dog waste under its stated terms.
iRobot's promise is a limited replacement programme, not proof that every accident will be avoided. It applies to solid cat or dog waste and is subject to its terms.
Eligible for strict must-haveExact-model independent test pass: RTINGS reports the E28 avoided pet waste, though it came dangerously close to contact.
+Exact-model independent test pass: RTINGS reports the E28 avoided pet waste, though it came dangerously close to contact.
Because clearance was close and carpet obstacle handling was weaker, RightRobot still presents this as risk reduction rather than certainty.
Eligible for strict must-havePromising variant evidence, but not strict-gate evidence: RTINGS says the tested Omni E28 is effectively the same robot as the E25 and it avoided pet waste, although it came very close.
+Promising variant evidence, but not strict-gate evidence: RTINGS says the tested Omni E28 is effectively the same robot as the E25 and it avoided pet waste, although it came very close.
RightRobot does not promote variant-equivalent evidence into a hard pet-waste must-have pass.
ECOVACS documents advanced AI obstacle recognition and an upgraded pet-care strategy, but RightRobot has not found an exact-model independent pet-waste pass strong enough for the strict gate.
+ECOVACS documents advanced AI obstacle recognition and an upgraded pet-care strategy, but RightRobot has not found an exact-model independent pet-waste pass strong enough for the strict gate.
MOVA explicitly says its binocular AI recognises pet waste and PetMate steers around it, but RightRobot has not found an exact-model independent pass yet.
+MOVA explicitly says its binocular AI recognises pet waste and PetMate steers around it, but RightRobot has not found an exact-model independent pass yet.
Xiaomi's manufacturer FAQ explicitly lists pet feces among recognised object categories, but RightRobot has not found a strong exact-model independent pet-waste test yet.
+Xiaomi's manufacturer FAQ explicitly lists pet feces among recognised object categories, but RightRobot has not found a strong exact-model independent pet-waste test yet.
Xiaomi also warns that recognition can vary with lighting, reflective/dark surfaces, viewing angle and sensor condition.
iRobot explicitly says PrecisionVision recognises and avoids pet waste.
+iRobot explicitly says PrecisionVision recognises and avoids pet waste. Vacuum Wars found its overall obstacle avoidance above average, but the published review does not provide a separate exact pet-waste test result strong enough for RightRobot's strict gate.
PrecisionVision is inactive during mapping/training runs, when iRobot says the floor should be clear of pet waste and cables.
iRobot explicitly says the Max 705's PrecisionVision AI recognises and navigates around pet waste, but RightRobot has not found a strong exact-model independent pet-waste test yet.
+iRobot explicitly says the Max 705's PrecisionVision AI recognises and navigates around pet waste, but RightRobot has not found a strong exact-model independent pet-waste test yet.
A feature claim is useful evidence of intended capability, but it is not enough for a no-contact must-have.
C28 uses LDS/laser-line obstacle avoidance rather than the camera-led system used on eufy's stronger pet-waste candidates.
+C28 uses LDS/laser-line obstacle avoidance rather than the camera-led system used on eufy's stronger pet-waste candidates. RightRobot has not found exact-model pet-waste evidence strong enough to verify it for this use case.
eufy's own current guidance says pet-waste detection should be treated as a safety layer rather than the only safeguard, and C28 instructions still tell owners to pick up small floor objects before cleaning.
Independent testing says pet-waste avoidance is not reliable enough for unattended use with accident-prone pets, especially when obstacle handling drops on carpet.
+Independent testing says pet-waste avoidance is not reliable enough for unattended use with accident-prone pets, especially when obstacle handling drops on carpet.
Independent testing says this model cannot identify or specifically avoid pet waste and should not be relied on for this use case.
+Independent testing says this model cannot identify or specifically avoid pet waste and should not be relied on for this use case.
Independent testing says the X10 Pro Omni does not always avoid pet waste, with obstacle avoidance notably weaker on carpet.
+Independent testing says the X10 Pro Omni does not always avoid pet waste, with obstacle avoidance notably weaker on carpet.
Independent testing says it avoids pet waste most of the time, which is useful but not strong enough for a hard no-contact requirement.
+Independent testing says it avoids pet waste most of the time, which is useful but not strong enough for a hard no-contact requirement.
Independent testing says the X11 OmniCyclone does not reliably avoid pet waste, so RightRobot will not pass it for a pet-accident must-have.
+Independent testing says the X11 OmniCyclone does not reliably avoid pet waste, so RightRobot will not pass it for a pet-accident must-have.
Independent testing finds inconsistent pet-waste avoidance, particularly on carpet, including runs where the robot bumps or drives into it.
+Independent testing finds inconsistent pet-waste avoidance, particularly on carpet, including runs where the robot bumps or drives into it.
Independent benchmark outside RightRobot's current core catalogue: RTINGS reports the Saros 10R is especially strong around pet waste and keeps its distance while cleaning around it.
+Independent benchmark outside RightRobot's current core catalogue: RTINGS reports the Saros 10R is especially strong around pet waste and keeps its distance while cleaning around it.
Do not transfer this result to the differently named Saros 10 or Saros 20; exact-model identity matters.
Eligible for strict must-haveIndependent benchmark outside RightRobot's current core catalogue: RTINGS reports consistent pet-waste avoidance on both hard floors and carpet.
+Independent benchmark outside RightRobot's current core catalogue: RTINGS reports consistent pet-waste avoidance on both hard floors and carpet.
This is useful benchmark evidence, but RightRobot does not substitute it for an untested different Dreame model or package.
Eligible for strict must-haveIndependent benchmark: RTINGS reports the X9 Pro Omni does not consistently avoid pet waste and can run over it on carpet.
+Independent benchmark: RTINGS reports the X9 Pro Omni does not consistently avoid pet waste and can run over it on carpet.
This is another example of why a strong general obstacle system cannot be assumed to be pet-waste safe.
See the independent obstacle-avoidance methodology RightRobot references →
No universal real-world guarantee is justified. Even strong independent test results reduce risk rather than eliminate it. iRobot offers a limited replacement promise on eligible j7/j7+ units for solid cat or dog waste, but that is a programme with terms rather than proof that every accident will be avoided.
Because a robot can avoid shoes and cables yet still miss pet waste. RightRobot records pet-waste evidence separately and requires exact-model independent support when a user makes it a hard requirement.
No. The strongest published tests RightRobot relies on use simulated solid pet waste. Wet or low-profile messes are a different detection problem and should not be assumed safe based on a solid-waste test.
RightRobot labels that as a manufacturer claim only. It can be useful context, but it will not pass the strict pet-waste must-have gate until stronger exact-model independent evidence is found.
A robot can gain capabilities and still become a worse buy for a particular household if those capabilities cost money without solving that household's problems. RightRobot keeps evidence, Home Fit and current shopping routes as separate layers so the recommendation can explain both fit and value.