Industries / Consumer goods / Baby products
Strollers and carriers
Strollers and carriers is the work of designing, making and testing the products that move a baby around, from prams, strollers (pushchairs), folding buggies and jogging strollers to the slings and soft carriers a parent wears.
7
Milestones achieved, from 1733 to 2018.
2
To watch for: announced, not yet confirmed outside.
3
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
1733
A goat-drawn baby carriage
William Kent made a small carriage for aristocratic children, shaped like a shell, decorated with snakes and fitted with springs, and pulled by a goat or small pony rather than pushed. Chiswick House's archive says Lord Burlington commissioned it for his oldest daughter in 1733 and calls it a forerunner of the modern pram; other accounts credit the 3rd Duke of Devonshire. The carriage belongs to the Duke of Devonshire and was shown at Chatsworth in 2024.
chiswickhouseandgardens.org.ukchatsworth.orguk.news.yahoo.com
Credit
William Kent
1840s
Prams made in numbers
By the Science Museum Group's account, perambulators, or prams, were mass-produced in Britain in the 1840s, and Charles Burton applied for the first patent in 1853. Queen Victoria bought three of his for the royal children, which boosted sales, especially among the upper classes. The V&A notes that 18th- and 19th-century pushchairs often had no simple brake or straps, that the writer Pye Henry Chavasse attacked them in 1839 as unsafe for young babies, who needed to lie flat, and that prams for lying flat did not arrive until wicker bassinet prams in the 1880s.
collection.sciencemuseumgroup.org.ukcollection.sciencemuseumgroup.org.ukcollections.vam.ac.uk
Credit
Charles Burton; Victorian pram makers
1965
A buggy that folds small
Owen Finlay Maclaren filed a British patent on 20 July 1965 for a baby carriage built on two X-shaped frames. Instead of folding flat, it collapses into a narrow bundle like a walking stick, which the patent calls "stick folding"; the US patent (3,390,893) followed in July 1968. Northampton Museums, which holds one made at Barby and bought in 1971, says its swivelling wheels were based on Maclaren's Second World War designs for aircraft undercarriages, and that the folded buggy hooks over the arm, leaving the hands free on public transport. Production stayed in England until the Long Buckby factory closed in 2009 and moved to China.
patents.google.compatents.google.comtimeline.northamptonmuseums.com
Credit
Owen Finlay Maclaren
2009
Hinges that cut fingertips
On 9 November 2009 Maclaren USA recalled about one million Maclaren single and double umbrella strollers, made in China, after 15 reports of children putting a finger in the hinge while the stroller was being unfolded, 12 of them fingertip amputations in the US. When the recall was announced again in May 2011 there had been 149 reported incidents in all, and 37 injuries, including five more fingertip amputations, had happened after the first recall.
Lessona recall reaches only some owners, and amputations went on for 18 months after this one.
Credit
Maclaren USA; US Consumer Product Safety Commission (CPSC)
2010
Sling deaths bring a rule
A sling is a length of fabric worn on the carer's body that holds a baby upright or reclined. On 12 March 2010 the CPSC warned that slings may pose a risk of suffocation; CBS News reported that the agency had found at least 14 deaths over 20 years, 12 of them babies younger than four months. Twelve days later Infantino recalled about one million SlingRider and Wendy Bellissimo slings in the US and 15,000 in Canada, after three deaths in 2009. ASTM International, a standards body, published a voluntary sling standard, ASTM F2907, in 2012, and the CPSC made it mandatory from 30 January 2018 (16 CFR 1228), after counting 16 deaths and 54 injuries from 2003 to 2013. The rule requires static and dynamic load tests and a warning that the baby's face must stay uncovered and visible and its chin off its chest.
Lessonmost deaths came from how the baby lay in the sling, not from a part breaking, so the rule pairs strength tests with warnings about position.
Credit
US CPSC; ASTM International; Infantino
2015
A mandatory US stroller rule
From 10 September 2015 every carriage and stroller sold in the US had to meet ASTM F833-13b, a voluntary standard first published in 1983, with one CPSC change to stop a baby's head being trapped at an adjustable grab bar. The agency's review counted 1,297 incidents from January 2008 to June 2013, including 4 deaths and 391 injuries, and found that hinges caused the highest injury rate of any stroller hazard. The 2013 editions of F833 had added a test for pinching at fold hinges: a gap fails if a 5.33 mm probe enters it but a 9.53 mm probe does not, since a finger can get in and be crushed. The CPSC counted 85 suppliers and about 3 million strollers sold to US households a year. Australia's mandatory standard, adopted in 2007, requires a red parking brake control and a tether strap with the warning "use this strap to stop the pram/stroller rolling away", and China requires strollers and baby walkers to have a China Compulsory Certification (CCC) certificate, issued by a designated body after testing in a designated laboratory.
federalregister.govlaw.cornell.edulegislation.gov.austc.group
Credit
US CPSC; ASTM International; Australian Government
2018
A jogging stroller lawsuit
A jogging stroller's front wheel is often held by a quick release, a lever clamp that lets the wheel come off without tools. In February 2018 the CPSC sued Britax, which had declined to recall, over about 493,000 BOB jogging strollers: about 200 owners had reported the front wheel coming off, letting the fork dig into the ground and tip the stroller, and at least 97 people were hurt, 50 of them children. The case settled in November 2018 without a recall. Britax agreed to an information campaign with an instructional video and offered owners a free modified part or 20% off a new stroller, without admitting a defect. In July 2024 the European standards body CEN published EN 1888-3, a separate standard for pushchairs used for running, jogging or skating.
Lessona hazard that depends on how the user fastens a part is hard to settle by recall, and this one ended in instructions and a video.
Credit
US CPSC; Britax Child Safety; CEN
Watch for
Strollers that brake themselves
Glüxkind, a Vancouver company, says its motorized strollers brake automatically when the carer lets go of the handle, adjust braking to the slope and park themselves when idle. Its Rosa model has two 500 W motors to help push and is listed from USD 2,290. BCBusiness reported a first production run of 150 units at the end of 2024 and shipments to eight countries; neither the article nor the product page cites an outside test of the braking. Europe's pushchair standard, EN 1888-1:2018+A1:2022, does not cover pushchairs propelled by a motor.
Achieved whenan independent lab or consumer tester publishes measured stopping distances and slope holding after the handle is released, with a weighted dummy in the seat, or a published standard covers powered strollers and the product is certified to it.
Claimed by
Glüxkind Technologies, September 2026
Europe rewrites its pram standard
CEN's working group on wheeled child transport began revising EN 1888-1, the European standard for pushchairs and prams, on 7 April 2026, with a draft planned for September 2026. The draft covers children up to 22 kg each, where the 2018 edition of Part 1 stopped at 15 kg and left heavier children to Part 2. Motor-driven pushchairs stay outside its scope.
Achieved whenCEN publishes the revised EN 1888-1 and national standards bodies adopt it.
Claimed by
CEN/TC 252/WG 3, through the German standards body DIN, April 2026
Open
01
Open brake and fold measurements
A bench rig and a public dataset of measurements for strollers bought at retail. For each stroller it records the slope at which the parking brake starts to slip, with sandbag loads of 15 kg and, for strollers rated for heavier children, 22 kg (the limits of EN 1888 Parts 1 and 2); the peak force needed to fold and unfold it; and whether each hinge gap passes the 5.33 mm and 9.53 mm probes of ASTM F833's pinch test. Only sandbags and weighted dolls ride in the seat.
The briefCredit · unclaimed02
Screening marketplace strollers
A listing screen and a test buy. First collect at least 1,000 stroller listings from Amazon, Temu, Shein and AliExpress in the US and UK, and ask a fixed set of yes/no questions of each listing's photos and text: does it show a full harness, does it name ASTM F833 or EN 1888, are warning labels visible, is a maker named. Then buy 30 strollers, 20 that the screen flags and 10 that it does not, and check each against the bench tests of the standards: the restraint holds a weighted doll, the parking brake holds on a 12-degree slope, and leg openings and hinge gaps pass the probes. Report how well the screen predicts real failures.
The briefCredit · unclaimed03
Carrier hip angles on manikins
An infant manikin with angle sensors in its hips, and a public table of the hip flexion (how far the thighs are raised) and abduction (how far the knees are spread) it settles into in at least 20 carriers, slings and hip seats. Each is compared with the spread-squat or "M-position" that the International Hip Dysplasia Institute (IHDI) calls healthiest, with the hips spread apart to the side and the thighs supported. No child is used.
The briefCredit · unclaimed
Open 01 · the brief
Open brake and fold measurements
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A bench rig and a public dataset of measurements for strollers bought at retail. For each stroller it records the slope at which the parking brake starts to slip, with sandbag loads of 15 kg and, for strollers rated for heavier children, 22 kg (the limits of EN 1888 Parts 1 and 2); the peak force needed to fold and unfold it; and whether each hinge gap passes the 5.33 mm and 9.53 mm probes of ASTM F833's pinch test. Only sandbags and weighted dolls ride in the seat.
Why it is still open
Consumer testers check these things but publish ratings, not measurements. Which? tests 30 to 40 pushchairs a year to BS EN 1888 and checks "whether they'll hold your pushchair still on a slope", but its method page gives no slip angles, forces or gap sizes. CHOICE in Australia rates 47 prams and strollers, and its safety tests "can only pass or fail". Consumer Reports tests "whether the brakes can hold the stroller in place on a steep incline", but its public coverage gives an angle only when a stroller fails, as when the Zoe Traveler "tipped at less than 12 degrees" in February 2026. In the 2014 US rulemaking a commenter asked for the brake test to be run on a 20-degree slope instead of 12; the CPSC kept 12. Searches found no public table of measured brake-slip angles or fold forces by model.
which.co.ukchoice.com.auclickondetroit.comfederalregister.gov
Built with what exists
A tilting plywood platform with a hand winch and a digital inclinometer; a digital force gauge, or a load cell with an HX711 amplifier on an Arduino or Raspberry Pi; machined probe pins 5.33 mm and 9.53 mm wide; sandbags and weighted dolls; a phone camera and OpenCV to mark the moment of slip; Python and a public Git repository.
First step
Build the tilt platform and test three strollers the team already owns, loaded with sandbags, to fix the method.
Done looks like
At least 20 models from at least 8 brands, each tested three times at each load, with repeat slip angles within 1 degree; fold and unfold forces in newtons; probe results for every hinge; drawings, code and CSV files published; a second team builds the rig from the drawings and matches three of the models within 1 degree.
Open 02 · the brief
Screening marketplace strollers
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A listing screen and a test buy. First collect at least 1,000 stroller listings from Amazon, Temu, Shein and AliExpress in the US and UK, and ask a fixed set of yes/no questions of each listing's photos and text: does it show a full harness, does it name ASTM F833 or EN 1888, are warning labels visible, is a maker named. Then buy 30 strollers, 20 that the screen flags and 10 that it does not, and check each against the bench tests of the standards: the restraint holds a weighted doll, the parking brake holds on a 12-degree slope, and leg openings and hinge gaps pass the probes. Report how well the screen predicts real failures.
Why it is still open
Regulators act on one product at a time, after it has sold. CPSC actions since late 2025 include a recall of about 15 convertible strollers sold on AliExpress whose restraints could fail (November 2025), and a June 2026 warning about B. Childhood strollers sold on Shein that broke the mandatory standard, whose maker did not respond. Which? has run test buys of cheap baby carriers and slings, 10 of 11 of which failed in 2022, and in 2025 matched almost 800 marketplace listings to 702 UK product safety reports. Searches found no test buy or listing screen for strollers.
Built with what exists
Browser Use 1.0 to gather listings where the sites' terms allow it (see the Note "Deleting 80% to start over"); Gemini Decision 1 to answer the yes/no questions on listing images and text in bulk (see the Note "A thousand questions, one pass"); the bench rig from the deliverable above; weighted dolls; SaferProducts.gov and the UK Office for Product Safety and Standards to report failures.
First step
Gather 100 US listings, write 20 questions, and score 50 listings by hand to check that the questions can be answered from a listing.
Done looks like
1,000 listings screened, with a person checking a random 100 and agreeing with the screen on at least 90%; 30 strollers checked at the bench; the share of flagged and unflagged strollers that fail at least one check, with the screen's precision and recall; every failure reported to the CPSC or the UK regulator.
Open 03 · the brief
Carrier hip angles on manikins
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
An infant manikin with angle sensors in its hips, and a public table of the hip flexion (how far the thighs are raised) and abduction (how far the knees are spread) it settles into in at least 20 carriers, slings and hip seats. Each is compared with the spread-squat or "M-position" that the International Hip Dysplasia Institute (IHDI) calls healthiest, with the hips spread apart to the side and the thighs supported. No child is used.
Why it is still open
IHDI says it "is not a certifying body" and "has never performed safety testing of any products". The published measurements are few. A 2023 ultrasound study of 13 infants compared one wide-base and one narrow-base carrier with a Pavlik harness, a brace used to treat hip dysplasia, and found that the narrow-base carrier lowered a measure of hip-socket shape, the Graf alpha angle, by 6.9 degrees. A 2021 Boise State student project put a newborn manikin of about 2.83 kg in three carriers, found a hand-held angle ruler hard to use, and proposed building a digital one into the manikin.
hipdysplasia.orghipdysplasia.orgeutils.ncbi.nlm.nih.govscholarworks.boisestate.edu
Built with what exists
An infant CPR or weighted training manikin fitted with rotary potentiometers or small motion sensors (IMUs) at each hip, read by an Arduino or Raspberry Pi; a small sensor board, which an AI model such as Claude Fable 5.2 can help design (see the Note "Three working devices"); a torso form to wear the carriers; photo checks with OpenCV; Python.
First step
Fit sensors to one manikin's hips and compare their readings with a hand goniometer in three carriers.
Done looks like
At least 20 carriers, each fitted three times by two different people at two manikin sizes, with repeat angles within 3 degrees; the manikin's angles in a Pavlik-harness reference position recorded as the baseline; a published table with a photo of every fitting.