Industrial manufacturing
Industrial Internet of Things | Industrial materials | Equipment Maintenance and Repair | Industrial programming |
home  MfgRobots >> Industrial manufacturing >  >> Industrial Internet of Things >> Sensor

KeepTruckin Launches AI‑Powered Dashcam with Ambarella CV22 SoC for Commercial Vehicle Safety

KeepTruckin has unveiled a next‑generation aftermarket dashcam that leverages Ambarella’s CV22 system‑on‑chip to deliver real‑time driver monitoring and advanced ADAS for trucks and other commercial vehicles.

Driver‑monitoring technology is now a cornerstone of automotive safety. While much of the conversation has centered on passenger cars, commercial fleets stand to benefit enormously from continuous driver oversight—reducing accidents, cutting insurance premiums, and giving fleet managers actionable risk data.

In partnership with edge‑AI vision SoC leader Ambarella, KeepTruckin developed a compact dashcam that uses a single CV22 chip to power a dual‑camera setup: a front‑facing camera for ADAS and incident recording, and an RGB‑IR camera for driver‑monitoring and recording.

The CV22’s high performance‑per‑watt design allows the device to run KeepTruckin’s proprietary AI algorithms locally, detecting high‑risk behaviors and issuing active warnings with minimal heat generation.

KeepTruckin Launches AI‑Powered Dashcam with Ambarella CV22 SoC for Commercial Vehicle Safety

“Driver safety has never been more critical,” said Jai Ranganathan, SVP of Product at KeepTruckin. “We chose Ambarella for its inference‑per‑watt efficiency and depth‑per‑lens capabilities, which were essential for our dual‑camera design.” Ranganathan added that the entire journey—from concept to market—spanned roughly 18 months, largely thanks to Ambarella’s comprehensive development partnership.

Abhishek Gupta, KeepTruckin’s Director of Product, noted the importance of Ambarella’s developer support, especially during a remote‑work pandemic.

With the CV22, KeepTruckin delivers a suite of AI features that identify high‑risk driving with high accuracy. The front camera flags close following, lane departures, forward collisions, speeding, and traffic violations. Simultaneously, the in‑cabin camera detects driver fatigue, distraction, and policy violations—including contextual cellphone use and seatbelt status—while integrating data from the front camera.

The dashcam streams pre‑processed video and still images to KeepTruckin’s Vehicle Gateway, which uploads them to the company’s cloud‑based fleet management platform in real time. The CV22’s image signal processor delivers 1440p HDR video across all lighting conditions, while on‑chip H.264/H.265 encoding keeps bandwidth and storage costs low.

KeepTruckin can push additional features to the CV22 via OTA updates, continually enhancing the platform. Its in‑house safety team refines model training in real time, shortening development cycles and improving precision.

“This is probably the most sophisticated AI vision product for automotive aftermarket,” said Udit Budhia, Ambarella’s Director of Marketing. “The CV22’s ISP supports multiple sensor types, HDR processing, and low‑light filtering, which reduce noise and improve video efficiency.” He added that Ambarella’s CVflow AI architecture allows multiple neural networks to run in parallel with low power consumption, boosting detection reliability.

Ambarella’s toolkit supports front‑facing and driver‑monitoring camera development simultaneously. The platform is fully compatible with industry‑standard frameworks such as PyTorch, ONNX, Caffe, and TensorFlow, making it easy for customers to port their own neural networks.

>> This article was originally published on our sister site, EE Times.


Related Contents:

For more Embedded, subscribe to Embedded’s weekly email newsletter.

Sensor

  1. Analog vs Digital Sensors: Types, Applications, and Practical Examples
  2. UltraSoC’s Bus Sentinel: Real‑Time Hardware IP Securing SoC Bus Transactions
  3. Reliable Arduino-Based RFID Attendance System Powered by Python
  4. Secure Arduino‑Based Keyless Door Lock with LCD Display and 4×4 Keypad
  5. Master Stepper Motor Control with Arduino & A4988 Driver – Step-by-Step Guide
  6. Advanced Vision System Breaks Through Clouds & Fog, Revealing Hidden 3D Scenes
  7. Ensuring Spectral Fidelity: Optimizing Hyperspectral Camera Quality
  8. Single‑Pair Ethernet: Reliable Industrial Connectivity
  9. Audio‑AI Stethoscope: Real‑Time Monitoring of Machine Health and Quality
  10. Leveraging AI and Robotics Across Multi-Inspection Systems for Flawless Production