Forklift pedestrian detection camera: warn when forklifts and people get too close
Shared aisles are where forklifts and people meet. Eye AI forklift and pedestrian safety is being built to watch those aisles through your existing CCTV, estimate the gap between a forklift and a person, and flag when it drops below a limit you set. Every close call is logged, so near misses become data you can act on, with no tags or sensors fitted to people.
ILLUSTRATION WITH SAMPLE DATA · IN PILOT
The problem
Near misses usually go unreported because nobody saw them or nobody wants to.
Without a record, you cannot tell which aisle or corner is the real hot spot.
Safety rules are hard to enforce when the evidence is a memory.
How it works
- Camera-readiness check.We look at your aisles and junctions and tell you which views suit proximity checking.
- Calibrate each camera on site.Distance is estimated from the camera view, so we set up each camera at your site and agree your minimum safe gap.
- Pilot.A computer at your site analyses the video, which stays on your network. During the pilot we review results with you.
- Review close calls.Each event shows the estimated distance, time and place on your password-protected dashboard.
What you get
- Proximity events with an estimated distance.
- A hot-spot map showing where near misses cluster.
- Shift reports for toolbox talks and reviews.
- Searchable history and CSV export.
- A motion safety net that shows movement the system could not name.
Where it is used
Honest limits
- Forklift and pedestrian safety is launching with pilot partners. It is in development and not yet generally available.
- We publish no accuracy figure now. Figures will be published after on-site validation, with the conditions stated.
- Distance is an estimate taken from the camera view. That is why each camera is calibrated on site, and why results differ between views.
- It is a review and reporting tool for your team. It does not stop a forklift and is not a replacement for physical barriers, mirrors, speed rules or driver training.
- Blind corners, racking and people hidden behind loads can hide a pedestrian from the camera.
- Phone alerts are part of the separate alerts add-on, which is also launching.
- Pilots are scoped to your requirements first: we agree the duration, the number of cameras and how success is checked on your own footage before anything starts.
Questions about forklift and pedestrian safety
Is this available today?
No. It is launching with pilot partners. Ask about the pilot if you want to take part.
Do forklifts or workers need tags?
No. It works from your camera views.
How is the distance measured?
It is estimated from the camera view and calibrated on site for each camera.
Can it stop a forklift?
No. It reports close calls so your team can act. It is not a vehicle control system.
How accurate is it?
We have no published figure yet. We will publish accuracy after on-site validation, with conditions.
Tell us about your site.
We read every request and reply within 24–48 hours (working days, Monday to Friday) with questions and a proposed solution.
Request the pilot