Reveal uploads video recordings when your vehicle tracker or AI Dashcam detects unsafe driving, such as hard braking. The AI Dashcam also monitors driving behavior to trigger in-cab audio alerts for drivers. These alerts help encourage safer driving habits over time.
Your company may be legally obligated to let drivers know what video data you capture. Share this article with your drivers so they know what information the camera generates, what is stored in Reveal, and how you use it.
In this article:
- Equipment required to detect unsafe driving
- How unsafe driving detection works
- Detecting events with a vehicle tracker
- Detecting events with an AI Dashcam
- Driver-facing camera setup and calibration
Equipment required to detect unsafe driving
For Reveal to detect unsafe driving, your vehicle must have one or more of the following:
- Vehicle tracker: A device installed in your vehicle to track movement and behavior.
- Built-in telematics solution: A built-in system provided by the vehicle manufacturer.
- AI Dashcam: To use video features, you must have an AI Dashcam installed in your vehicle.
How unsafe driving detection works
| Feature | Vehicle tracker | Cloud AI | Camera AI |
|---|---|---|---|
| What type of AI does it use? | None | Cloud AI | Camera AI |
| Where is it? | On the tracker | On a remote server | On the AI Dashcam |
| What does it do? | Tracker triggers a harsh driving event based solely on factors such as speed and g-force. | Cloud AI, also known as the video analytics engine, analyzes the severity of an event after it has happened. It also recognizes incidents such as Tailgating. | Camera AI constantly monitors driving behavior, provides immediate feedback to drivers and triggers videos to be uploaded to Reveal. |
| When does it run? | After the harsh driving event. | After the event. It analyzes videos moments after it receives them from the camera. | In real-time. It runs constantly, when the vehicle is on, to monitor driving behavior. |
| How does it affect classification? | Generates events | Classifies events | Generates event |
Detecting events with a vehicle tracker
The following unsafe driving behaviors are detected by the vehicle tracker:
- Hard acceleration: The vehicle speeds up suddenly (by more than 5 mph (8 km/h) per second). The g-force is above 0.220g for at least 1 second.
- Harsh cornering: The vehicle makes a harsh turn at speed. The g-force is above 0.4g for at least 2 seconds.
- Hard braking: The vehicle slows down suddenly (speed decreased by more than 6 mph (9.5 km/h) per second). The g-force is above 0.265g for at least 1 second.
- Sudden force: The vehicle jolts suddenly (caused by impact with a person, animal, or object, for example). The g-force is above 0.7g for 0.1 seconds. Sudden force is only detected by certain device models. Contact Support to find out whether your device can detect it.
Detecting events with an AI Dashcam
Driver-facing triggers
The driver-facing camera monitors a driver’s head position and object use, such as cell phones. If the system detects unsafe behavior, it triggers an in-cab alert. The AI then classifies the event and uploads a video to Reveal.
Driver-facing triggers include:
- Tiredness
- Distraction
- Smoking
- Phone call detection
- Seat belt unfastened
- Food in hand
- Driver camera covered
Road-facing triggers
The type of lens that the camera is fitted with can affect how close and how fast objects appear. The road-facing camera monitors driving behavior and the vehicle's proximity to other road users.
Road-facing triggers include:
- Tailgating
- Pedestrian collision warning
- Solid line crossing
- Rolling stop
- Traffic light violation
- Posted speed sign
- Driver camera covered
Driver-facing camera setup and calibration
The camera monitors the driver's head position during the first 30 seconds of a journey and every 30 minutes after that. This setup process ensures the AI accurately detects unsafe behavior.
The AI setup can be disrupted if the driver:
- Wears glasses, a hat, or a mask.
- Frequently touches their face.