Skip to content
IEMI Labs

ForkGuard AI

Retrofit forklift camera that warns drivers before they hit someone

Concept · not for sale
Engineering design: ForkGuard Log
Target price from$799per forklift (rear camera option)

Software: $15–$25 per month per forklift

Concept, not for sale yet. Tell us if you want it; that decides what we build.

Rate this idea and tell us what to change
ForkGuard AI concept: the idea and where each part goes

Drag to rotate · right-drag to move · scroll to zoom · click a part for its size and placeSwipe to rotate · pinch to zoom · two fingers to move · tap a part for its size

How it fits together

ForkGuard AI, part by part

Scroll or tap a part. Press “See the 3D CAD design” to watch the model, built from our CAD, light each part up and show how it connects.

ForkGuard AI concept: the idea and where each part goes

Real-time 3D render · drag to rotate · final design may varySwipe sideways to rotate

  1. 01

    The whole ForkGuard AI

    A camera on the forklift guard spots people behind the truck and chimes the driver in under 300 ms; an impact sensor logs rack hits.

  2. 02Brain

    Raspberry Pi Zero 2 W

    The brain: it runs the fg-vision, fg-imu, fg-io program

  3. 03

    AI Camera (IMX500)

    Finds people behind the truck, on the sensor itself

    Raspberry Pi Zero 2 W · CSI-2 ribbon cable · Person boxes, no images

  4. 04

    Impact sensor

    Feels hits at 400 Hz and wakes the logger

    Raspberry Pi Zero 2 W · I²C and GPIO 4 interrupt · Acceleration, impact wake

  5. 05

    Cab LED bar

    Shows ready, check due or a red warning flash

    Raspberry Pi Zero 2 W · GPIO 18, level shifter · LED colour data

  6. 06

    Cab piezo chime

    Gives a three-beep warning to the driver

    Raspberry Pi Zero 2 W · GPIO 13 PWM and MOSFET · 2 kHz chime drive

  7. 07

    Cab mute button

    Driver presses to mute the chime for 10 s

    Raspberry Pi Zero 2 W · GPIO 27 pin · Button press

  8. 08

    Truck power converter

    Turns truck battery power into 5 V for the unit

    Raspberry Pi Zero 2 W · 5 V on M12 cable · 5 V power

  9. 09

    All connected

    7 key parts, one program, and your phone

What it is

A rugged camera for the forklift roof with an in-cab display that warns of pedestrians.

What it does

  • Detects people near the forklift
  • Beeps and flashes in the cab
  • Logs anonymous near misses
  • Shows hotspots on a dashboard

Who it's for

Warehouses, 3PLs, factories, builders' merchants and retail distribution centres.

The idea

One small kit on each forklift: impacts, reversing near misses and the daily pre-shift check, on one fleet dashboard

The problem today

Small warehouses with 5 to 40 mixed-brand forklifts find rack and door damage days later, with no idea which truck or shift did it. Near misses behind a reversing truck are never written down, so nobody sees the hotspot until someone is hurt. The OSHA daily pre-shift check is a paper sheet that is often skipped or filled in at the end of the week.

With ForkGuard AI

A camera unit on the back of the overhead guard and a cab pod log every impact and every person close behind a reversing truck, and chime in the cab as a driver aid. The driver scans a tag on the truck to do the pre-shift check on a phone. The manager sees impacts, near-miss hotspots and check compliance for every truck on one dashboard. It never slows or stops the truck.

Rack aislesPick-face walkwayDock doorsCharging bayShift office1234
  1. 1Camera unit on the overhead guardLooks back and down from the rear crossbar: logs each person close behind while reversing and every bump, with the truck, time and place
  2. 2Cab pod beside the driverChimes and lights when someone is close behind while reversing, as a driver aid only; the driver can mute it for a set time
  3. 3Check tag on a truck in the charging bayThe driver scans it at the start of the shift and does the daily pre-shift check on a phone, with a photo for any fail
  4. 4Manager's dashboard in the officeImpacts the same shift, near-miss hotspots such as an aisle end on the walkway, check compliance and hours for every truck
A small warehouse floor: rack aisles, a pick-face walkway at the aisle ends, dock doors, the charging bay and the shift office, drawn to show where each part goes (not to scale).

How the parts connect

ForkGuard AI, wired up

Every part, what links it to the brain, and how the data reaches your phone. A solid line is a cable; a dashed line is a radio link.

AI Camera (IMX500)Person boxes, no imagesCSI-2 ribbon cableImpact sensorAcceleration, impact wakeI²C and GPIO 4 interruptCab LED barLED colour dataGPIO 18, level shifterCab piezo chime2 kHz chime driveGPIO 13 PWM and MOSFETCab mute buttonButton pressGPIO 27 pinTruck power converter5 V power5 V on M12 cableTHE BRAINRaspberry Pi Zero 2 Wruns fg-vision, fg-imu, fg-ioNode.js servicesSite broker or cloudvia Wi-Fi · MQTT/TLSSupervisor phonevia Web app and SMS
Drawn from the ForkGuard AI parts list. The 3D view above is built separately, from the CAD.

The program

What controls ForkGuard AI

Runs on
Raspberry Pi Zero 2 W
Written in
Node.js services
System
Raspberry Pi OS Lite, read-only root
AI model
Person detector on the IMX500 camera chip
  1. 1Read person boxes from the camera and track them
  2. 2Turn each foot point into a distance band behind the truck
  3. 3If reversing, moving and someone is close, chime and flash red
  4. 4Log impacts over 3 g from the accelerometer
  5. 5Store events in SQLite, then send them over Wi-Fi
fg-vision, fg-imu, fg-io.tsmain loop
camera.on('detections', boxes => {
  const people = tracker.update(boxes.filter(b => b.score >= 0.45));
  const bands = people.map(p => distanceBand(p.footPoint));   // danger, close, near
  const near = bands.some(b => b !== 'near');
  if (io.reverse && motion.moving && near) {
    if (++closeFrames >= 3 && !muted) io.chime(), io.led('red-flash');
  } else closeFrames = 0;
  if (near) db.insert('nearmiss', { band: closest(bands), zone });
});
imu.on('sample', a => {
  if (highPass(a) >= 3.0 && lasted(5)) db.insert('impact', peak());
});
setInterval(() => agent.flush(db, mqtt), 1000);   // when Wi-Fi is up

To your phone

How the data reaches ForkGuard dashboard and check app

Supervisor phone or laptop browser

  1. Truck unit
  2. Site broker or cloud
  3. Supervisor phone

What you see in the app

One screen for each job ForkGuard AI does.

  • Reversing near missFL-07 backs out of aisle 14 as a picker steps behind it. The cab chimes and flashes red as a driver aid, and the near miss is logged.

  • Impact logA truck clips a rack upright. A minute later the supervisor has a text, so the damage is found and checked the same shift.

  • Pre-shift checkBefore each shift the driver scans the truck's QR tag and works through the daily check on a phone. A truck with no check shows amber.

App screens with sample data. The interface is in development.

What the app gets

Data ForkGuard AI sends to the app
Impact event4.1 g, majorEach hit
Near-missClose band, aisle 14Each reverse event
Shift summary266 min moving, 1 impactEach key-off
Pre-shift checkPass, FL-07Each shift
Unit healthREADY, 12 fpsEvery 5 min

Video is not recorded or sent. Only events, with the truck and zone, are stored.

CAD design · Design complete (stages 1-6), not built

Designed in code, checked in 3D

  • 10 printed parts, each a single closed solid
  • Every bought part fitted inside and checked for collisions
  • Assembled exactly as shown in the 3D CAD view (the button on the picture at the top)
See the CAD parts renderForkGuard Log printable parts rendered from the CAD

Key specs

ForkGuard AI hardware
AI processorHailo-8L (13 TOPS) + ARM host in the camera unit
CameraWide-angle 1080p camera (~150° HFOV) with IR illumination for dark aisles; optional rear camera
SensorsIMU for motion and impact detection · Ambient light sensor
ConnectivityWired link to cab display · Wi-Fi 6 for near-miss upload · Optional LTE · Digital output for truck speed-limit input (where supported)
Power12–80 V DC forklift supply with isolated converter, ~6 W (target)
EnclosureBlack rugged IP67 camera unit on overhead-guard clamp; in-cab display with buzzer and LED bar
DimensionsCamera unit 14 × 8 × 6 cm; cab display 10 × 7 × 3 cm
WeightCamera ~600 g; display ~200 g
Ingress ratingIP67 camera unit; IP54 cab display
All target specifications

Detection

Detection target specifications
Field of view~150° horizontal
RangeUp to ~10 m
Warning latencyUnder 200 ms (target)
Low lightIR illumination, works in dark aisles

Alerts

Alerts target specifications
Cab displayDirection and zone indicator
BuzzerUp to 85 dB, adjustable
OutputDigital out for speed-limit input
EventsAnonymous near-miss log

Power and physical

Power and physical target specifications
Supply12–80 V DC, ~6 W
Camera unit14 × 8 × 6 cm, IP67
Cab display10 × 7 × 3 cm
Operating temp.−25 °C to 50 °C (freezer option planned)

Targets for production units; they may change during development.

Pricing

Hardware: $799–$1,099 per forklift (rear camera option)

Software: $15–$25 per month per forklift

Target pricing. Fitting by your forklift service provider or our partner network is quoted per site.

Warn

$799 + $15/mo per truck

Forward pedestrian warning with near-miss log.

  • Forward camera
  • Cab display and buzzer
  • Near-miss dashboard
  • Monthly reports

Warn + Rear

$1,099 + $25/mo per truck

Adds a rear camera and speed-limit output.

  • Forward and rear cameras
  • Speed-limit output
  • Heat maps
  • Weekly reports

Fleet

Custom

For multi-site fleets and rental companies.

  • Volume pricing
  • Telematics integration (planned)
  • Fitting partners
  • SLA

Indicative launch pricing in USD, excluding taxes and shipping. Final pricing is confirmed before any balance is charged.

Community

What do you think of ForkGuard AI?

ForkGuard AI is in development. Rate the idea, tell us what to change about the features, specs or price, and we'll build it with your feedback.

How much people want it

–

No ratings yet

  1. 50
  2. 40
  3. 30
  4. 20
  5. 10

Follow ForkGuard AI

Get build updates and first access when pre-orders open.

Leave feedback

Your rating of the idea

Feedback on a product concept in development, not a review of a shipped product. Posts may be moderated; links and contact details are held for review. See our community guidelines and privacy policy.

Comments

    No comments yet. Tell us what you'd want from ForkGuard AI.

    Questions

    Ask about ForkGuard AI

    Tell us what this would save you, or ask anything. We reply within two business days.

    We use your details only to respond to this request. See our privacy policy.

    From the product lab · ideation phase

    Lab updates for ForkGuard AI

    All lab updates
    1. LatestSmall warehouses with 5 to 40 mixed-brand forklifts find rack and door damage days later, never record near misses behind reversing trucks, and keep the OSHA daily pre-shift check on paper (OSHA 1910.178(q)(7), sources in DESIGN.md).
      • Drawn idea plan (idea.json): a small warehouse floor from above with the rack aisles, the pick-face walkway, dock doors, the charging bay and the shift office; pins for the camera unit on a reversing truck, its cab pod, the check tag on a truck in the charging bay and the manager’s dashboard
      • The product picture is now this drawn plan instead of the plain fallback card; drawn from the design notes, not the CAD
      • What the plan keeps from the market work: it sells evidence, not protection; the cab chime is a driver aid only and it never slows or stops the truck

      See it in 3DFull research and design

    2. On a waste tipping floor three problems meet: lithium batteries hidden in loads start fires (UK waste firms reported 541 battery fires in the year to March 2026, waste360/ESA), loaders reversing among people on foot lead to big HSE fines (London Energy, £400,000 in 2025), and wrong items in a load cost money unless there is a photo to back-charge the customer (3rd Eye, WM). Our own three products for one hall cost $6,286 and $163 a month.
      • Combined into RecycleYard (https://iemilabs.com/products/recycleyard/); ForkGuard AI itself is unchanged

      See it in 3DFull research and design

    3. A builders’ merchant, timber yard or garden centre lets customers drive in, park, queue at the will-call counter and walk among forklifts. WorkSafe NZ fined a timber yard after a forklift knocked a pack onto a customer and asked yards to control people on foot and show customers where to park. Today each job is a separate tool: about $950 a truck for an AI reversing camera, $35 a month per camera for plate reading, paid processing for traffic counts and $859-919 per people counter.
      • Combined into SafeYard (https://iemilabs.com/products/safeyard/); ForkGuard AI itself is unchanged

      See it in 3DFull research and design

    ForkGuard AI FAQ

    General FAQ
    Does ForkGuard identify or record people?

    No. Detection runs on the camera unit and video is not stored. Near-miss events are anonymous: they record time, truck, distance and location, not who was involved. Optional snapshots are blurred on the device before upload.

    Is ForkGuard a safety-rated system?

    No. ForkGuard is a driver-assistance warning aid. It does not replace training, traffic management, segregated walkways or the safety measures your regulator expects. Safety-function certification is not claimed.

    How is it installed?

    The camera clamps to the overhead guard, the display mounts in the cab and the harness connects to the truck's 12–80 V supply. Fitting takes about an hour per truck and can be done by your forklift service provider.

    What connectivity does it need?

    Warnings work entirely on the truck with no network. Near-miss events upload over site Wi-Fi when available, or over the optional LTE module.

    Is ForkGuard AI available now?

    No. It is a new concept in development. Specifications are target specifications. It is a concept: we are not taking orders, reservations or deposits, and it has no ship date yet.

    What warranty will it come with?

    We are planning a 2-year limited hardware warranty covering manufacturing defects. Final terms will be published before shipping begins.

    Related products

    ParkSense AI concept: the idea and where each part goes
    VisionConcept

    ParkSense AI

    A palm-sized edge box that plugs into your existing parking cameras.

    • Counts free bays in real time
    • Flags overstays and wrong parking
    • Guides drivers to open spaces

    From $499 per edge box

    Details
    WasteSort Coach concept: the idea and where each part goes
    VisionConcept

    WasteSort Coach

    A clip-on camera coach for the recycling bins you already have.

    • Lights the right bin for each item
    • Says "not sure" instead of guessing
    • Estimates each station's contamination

    From $299 per 3-bin station

    Details
    IEMI Edge Hub + VisionCam concept: the idea and where each part goes
    VisionConcept

    IEMI Edge Hub + VisionCam

    An on-site AI box for up to 8 cameras that sends events, not video, starting with parking.

    • Counts free parking spaces live
    • Texts staff about overstays
    • Keeps video on site

    From $1,190 LotWatch installed kit

    Details

    ForkGuard AI

    Concept · target $799

    I want this