RIDE INTELLIGENCE PLATFORM

Ride Intelligence for micromobility

We turn every ride into data: a continuous and detailed record of what happened and where. On any hardware stack, even your smartphone.

Get in touch Explore open hazard data

A lineup of light vehicles: bicycle, cargo bike, delivery trike, delivery robot, enclosed e-bike, e-scooter and e-moped.

Every other mode of transport is measured. Light vehicles are not.

Aircraft, trucks and cars are measured continuously. Every trip leaves a record that can be checked against a standard.

Light vehicles such as bicycles, cargo bikes, e-scooters and e-mopeds now carry a serious share of urban traffic and urban logistics. Almost none of it is measured. Routes, traffic flow, risk profiles, speeds and rule adherence remain largely unknown.

BlossomEdge closes that gap, without any new hardware on the vehicle.

Get in touch

Isometric city map showing light vehicles such as bicycles, cargo bikes and scooters travelling along highlighted routes

One measurement stack. Multi-level insights.

01 CAPTURE

On-device AI detects relevant information and records video, sensor input, location and time in real time.

02 FUSE

BlossomEdge data combines with public and client data.

03 INTERPRET

Cloud AI reads the combined record and generates critical insights and signals.

04 OUTPUT

Dashboards, reports, behaviour analysis, API events for downstream systems and apps

The same ride, read at multiple levels

We generate insights on several levels of abstraction to inform decisions or drive automations in downstream systems.

Gyroscope

Accelerometer

Magnetometer

Camera

GPS

LIDAR

...

Raw Sensor Data

Use cases enabled by our solution

01 / 10

Riders and events

Ride/r risk scoring

Quantifying how risky or safely a person or vehicle rides.

Risk-scoring riders and rides based on behaviour, events, and external conditions

Claims handler reviewing incident evidence

Riders and events

Claims handling

Evidence that settles what happened.

A record of what happened, where and when, instead of two conflicting accounts.

Rides and vehicles

Cargo delivery certification

Evidence of transport quality, end to end.

Proof that a transport met a defined quality standard, from pickup to arrival.

E-moped display showing a hazard warning

Rides and vehicles

Light vehicle ADAS

Warnings the vehicle gives while riding.

Real-time hazard warning inside an e-bike, a scooter or a moped.

Rides and vehicles

Autonomous driving

Risk perception for light autonomous vehicles.

Real-time risk perception for PDDs and other light autonomous vehicles as input to their self-driving stack

Streets and networks

Infrastructure planning

Evidence for where to build and proof of what it changed.

Hazard evidence before a crash happens, and a measure of what actually changed after a road was rebuilt.

Handlebar-mounted phone recording a street

Riders and events

Dash cam

Incident footage, in an app or a device.

Continuous recording that keeps the seconds around an incident. As an app, or built into hardware.

Routing view on a city map with hazard points

Streets and networks

Safe routing

Routing that accounts for risk, not only distance.

We tell you where the dangerous spots are, so you can route a rider, a user or a cargo around them.

Riders and events

Rider training & feedback

Feedback for beginners, delivery riders, other professionals, and children

Feedback on how someone actually rides. For children learning the route to school, for adults returning to a bike, and for professional riders.

Streets and networks

Research

Field data for road safety and mobility research.

Structured field data on risk, behaviour and infrastructure, collected under real conditions.

Do you see yourself in these use cases? We would like to hear from you.

Get in touch

Built ride by ride

All street level data is backed by video, post-processed for QA, and labelled against risk taxonomies. Collected across several cities and several kinds of rider.

None of it can be reconstructed after the fact from GPS traces or crash reports.

34.241

hazards detected

cities

15.233

km analysed

active projects

Hazard heat map of Munich showing risk hot spots

Explore open hazard data

Running projects

Each runs with a named partner organisation.

LMU München, Social Geography

Field study with Prof. Dr. Henrike Rau of LMU's Professorship in Social Geography on how the risk riders feel relates to the risk the models measure.

Read the project

Radentscheid München

Rider-collected hazard evidence for infrastructure decisions in Munich.

Read the project

QuadRadEntscheid Mannheim

Creating a cycling hazard map for Baden-Württemberg's 2nd largest city

Read the project

London hazard map with clip review panel

London Cycling Campaign

UK field study collecting hazard data with a local cycling organisation.

See the data

Further work runs with Technical University Munich and with partners in Amsterdam, Cambridge, Augsburg, and Cologne.

Go-Bike

Go-Bike is our own app and the place the models meet the road. Riders mount a phone, get a warning when a situation turns dangerous, and review what happened afterwards. It is free, it is in both app stores, and is being used in multiple cities across Europe.

Go-Bike app screens showing hazard maps, ride recordings and ride history

Built around data we are trusted with

Riders hand us video of the streets they live on. That sets the standard for how the system is built. We follow the European GDPR in all countries worldwide.

Processing on the device

Detection runs locally while riding. No continuous video stream leaves the phone.

Anonymised before it is shared

Faces and number plates are blurred. Route starts and ends are cut, so a trip cannot be traced to a front door.

Consent per ride

Riders decide what is shared and with whom, ride by ride, and can withdraw later.

Documented legal basis

Built against UK and EU GDPR, with published terms naming every category of data and every recipient.

The people behind it

Three founders who combine quantitative cognitive psychology and production machine learning with experience in building perception and decision solutions for mobility.

Dominic Noy

Technical

PhD in road-user modelling with multiple publications and patents Built perception behind ADAS, autonomous driving and smart glasses for Nissan, Meta, others Ex Director of Data Science at Humanising Autonomy

Dominik Busching

Product

Experienced Product Lead and Business Strategist, Diploma in Cognitive Psychology Built winning B2B SaaS and Consumer Products that scaled to millions of customers Ex Deloitte Strategy, ex tado°

Ron Pelley

Commercial

Serial founder and business development executive in wireless, telecom, automotive and AI Many years of experience in commercialising new technologies Commercial leadership for Humanising Autonomy, Soma Networks, Flarion, Qualcomm, SpiderCloud, and others