Cut through the noise with AI-powered video analysis

For many medium and large fleets using video telematics, the manual review of video footage is simply not scalable when triggered events can exceed hundreds per day. With NARA (Notification, Analysis and Risk Assessment), you can automatically remove false positive events and monitor driver behavior to enhance your duty of care, improve driver and road user welfare, and reduce associated costs.

Get in touch

NARA

Ground-breaking computer vision model to automatically remove false positive events

NARA is device agnostic so can be integrated with existing connected camera technology – whether VisionTrack or third-party hardware – making it ideal for any new or existing video telematics solution. The software also adds another powerful layer of analysis to AI vehicle cameras, installed with edge-based AI technology, that are often limited by the processing capacity of the device.

NARA is changing how collisions, near misses and driving events are analyzed, monitored and reviewed by automatically analyzing footage without the need for human involvement.

Identify and remove false positives

Effectively monitor driver behavior

Achieve rapid risk and FNOL intervention

Conclude fault claims settlements within 24 hours

Enhance duty of care commitment

Quickly validate driver welfare

Support road safety and driver coaching strategy

Helps deliver claims savings ($2,400 on average) for each collision detected

Identify and remove false positives

Effectively monitor driver behavior

Achieve rapid risk and FNOL intervention

Conclude fault claims settlements within 24 hours

Enhance duty of care commitment

Quickly validate driver welfare

Support road safety and driver coaching strategy

Helps deliver claims savings ($2,400 on average) for each collision detected

1. Event triggered 1. Event triggered
by a connected dashcam’s 100Hz accelerometer.

2. Video
automatically
uploaded 2. Video automatically
uploaded to the
Autonomise.ai
IoT platform.

3. Advanced object recognition software 3. Advanced object
recognition
software uses deep
learning algorithms
to identify different
types of vehicles,
cyclists and
pedestrians, and
their proximity at the
point of the
suspected incident.

4. Detects, tracks
and alerts on
events 4. Detects, tracks and
alerts on events with
accuracy levels
exceeding 98% where
the incident occurs in
view of the camera.

5. NARA
provides an
added layer of
analysis 5. NARA provides an
added layer of analysis
to determine if a
collision has occurred.

6. Real-time
alerts 6. Real-time
alerts for fleet
managers and
FNOL teams.

Event triggered

by a connected dashcam’s 100Hz accelerometer.

Video automatically uploaded

to the Autonomise.ai IoT platform.

Advanced object recognition software

uses deep learning algorithms to identify different types of vehicles, cyclists and pedestrians, and their proximity at the point of the suspected incident.

Detects, tracks and alerts on events

with accuracy levels exceeding 98% where the incident occurs in view of the camera.

An added layer of analysis

to determine if a collision has occurred.

Real-time alerts

for ﬂeet managers and FNOL teams.

How it works

NARA distinguishes between driving incidents and false positives to provide a fast and highly accurate AI solution to the fleets and their insurance partners. This significantly reduces the number of incidents that require manual validation and eliminates the issue of human availability or error.

Video
Telematics

Data
Analysis

Filter out
false positives

Real-time
Alerts

Driving AI Innovation

Using the latest in AI, computer vision and machine learning

NARA uses some of the latest advances in AI, computer vision and machine learning. This provides commercial fleets with accurate incident validation and supports proactive risk intervention.

AI Analysis

Providing an added layer of analysis

Following a video upload to the Autonomise.ai IoT platform – triggered by a connected camera device – NARA will determine exactly what has occurred. This complements a camera’s 100Hz accelerometer, which captures accurate speed and g-force data at a frequency of 100 times per second.

Rapid Validation

Automatically validate footage in eight seconds

NARA takes just eight seconds to automatically validate footage of collisions and near misses using advanced object recognition to identify nearby road users and their proximity at the point of the suspected incident and, if necessary, generate a high-priority alert. With a high degree of speed and accuracy, it is possible to eliminate false positives that are generated by harsh driving, potholes or speed humps.

Object Recognition

Advanced object recognition with 99% accuracy

Advanced object recognition software uses deep learning algorithms to train Autonomise.ai to automatically identify different types of vehicles, cyclists and pedestrians. With accuracy levels already exceeding 99%, where the collision occurs in view of the camera, the platform detects, tracks and alerts on actual events.

Driver Welfare

Immediately identify if a driver needs support

NARA includes Occupant Safety Rating (OSR) that translates forces experienced within the vehicle to calculate the percentage probability of injury, so fleets can immediately identify if a driver needs support. This early warning system provides a safety net that saves lives and minimizes driver distress.

NARA in action

NARA is already supporting fleet customers and industry partners to make a huge difference in terms of road safety, driver welfare and claims management.

Fleet

Fleet Customers

One of the UK’s leading logistics providers is using NARA to dramatically change how video footage is analyzed for a 1,100-strong vehicle fleet that normally covers over 46,000 journeys and 1.3 million miles a week.

The company generates over 200,000 weekly g-force events, of which, 99% are dismissed automatically as false positives by VisionTrack’s advanced AI algorithms. This leaves, on average, 2,800 priority videos that would have previously taken over twelve hours to review. Using NARA, the events needing human validation and intervention reduced to 15, making it possible to check in a matter of minutes.

The software also detected 400 near misses with other motor vehicles and 90 near misses with vulnerable road users, which is also supporting the company’s road safety and fleet risk reduction strategy.

Partner

Industry Partners

sopp+sopp, the leading accident management company, teamed up with VisionTrack to launch an auto FNOL service for commercial fleets, which is underpinned by the power of NARA. The solution is helping to validate driver welfare, streamline claims handling and reduce insurance costs.

“Auto FNOL is going to play an important role moving forward, helping our fleet customers to take control of the claims management process and support their duty of care commitments. We will gain faster, more accurate reporting, enabling us to reduce unreported incidents and safeguard driver wellbeing. The technology will also achieve cost savings through improved third-party capture rates and liability decision-making, including the evidence needed to challenge alleged and fraudulent incidents.” Callum Langan, Commercial Director, sopp+sopp.

Empowering the latest AI technologies

We are using increasingly sophisticated AI technologies to automate management processes, data analysis and incident detection in a wide range of fleet, insurance, road safety and risk management settings. Ground-breaking computer vision and machine learning innovation is enabling Autonomise.ai to automatically detect, track and alert on situations involving vehicles, so fleets can make best use of their video telematics and gain optimum value.

Find out more