The challenge:
a video telematics platform that no longer keeps up with the demand
VisionTrack is a leading global provider of AI video telematics and connected fleet data solutions. The company’s products transform road safety, claims management, fleet compliance, and operational risk.
As VisionTrack’s customer base increased, its video telematics platform also started to face growing challenges.
Built on non-scalable solutions, the original system struggled to accommodate the skyrocketing number of requests. As a result, customers experienced availability issues, while the data integration system suffered from inefficiencies, which hampered its performance and reliability.
Another vital issue stemmed from vendor lock-in. The original system worked only with the software from the device producer (Streamax). Its operation was limited to Windows, restricted the number of served devices, and would quickly use up all operational memory. All ad-hoc system amendments and maintenance were costly. Recognising the need for a modern and scalable system, VisionTrack partnered with Internetum, an experienced European IoT developer.
Analysing the requirements & potential solutions
Working hand in hand, the VisionTrack and Internetum teams analysed the existing infrastructure conditions and needs.
They identified the following areas for improvement:
#1: Scalability
The legacy video telematics platform could not adapt dynamically to fluctuating workloads, leading to frequent service disruptions.
#2: Integration
Device integration relied on bespoke, fragmented implementations, which were inefficient and challenging to maintain.
#3: Testing
Real-world testing of telemetry devices required physical vehicles, consuming significant resources and time.
The team established that the platform requires a robust cloud infrastructure with a boost in the form of AI analytics and advanced testing mechanisms.
Core aspects of the upgraded platform include:
VisionTrack adopted a comprehensive video telematics solution built on the Microsoft Azure ecosystem.
Core aspects of the upgraded platform include:
-
Refactored integration architecture
Device-agnostic data platform
Internetum developed a new, independent integration layer that handles data from any device manufacturer. This approach ensured seamless integration with hardware from DTEG, Streamax, and others, including ATrack, Howen, Aplicom, and the JT/T protocol for multi-vendor compatibility.
Scalable microservices
By migrating from a monolithic architecture to microservices, VisionTrack unlocked independent scaling of the system based on varying traffic demands. Internal APIs facilitated smooth communication between these services.
-
Scalable infrastructure with Azure functions
The video telematics solution uses Azure’s serverless computing capabilities for event-driven scalability and automatically adjusts them to meet the demands of live telemetry data. As a result, the platform ensures high availability and operational efficiency, even during peak loads.
-
AI-enabled analytics
Internetum enriched VisionTrack’s video telematics platform with multiple AI mechanisms for analysing road events. This approach allowed refining the company’s proprietary solutions and improved event detection accuracy and response times.
-
Advanced testing mechanisms
Hardware emulation
Internetum introduced hardware emulators for GPS signal simulation and virtualised camera outputs, significantly reducing the reliance on field testing.
Video-based testing
Controlled video inputs enabled consistent testing of edge cases like dangerous driving in a safe and repeatable environment.
Device emulation
A custom software emulator for Streamax devices allows playing back recorded routes from test and production environments. This innovation facilitates efficient debugging and optimising event detection and classification algorithms.
Results: a future-proof video telematics platform
The implementation of Internetum’s custom software solutions delivered tangible benefits for VisionTrack:
A01
Cost reduction
Scalable infrastructure and virtualised testing mechanisms reduced infrastructure and testing costs by eliminating the need for extensive physical resources.
A02
Enhanced scalability
The microservices architecture and Azure-based infrastructure allowed VisionTrack to scale services dynamically, accommodating new clients and fleets effortlessly.
A03
Optimised performance
AI-powered analytics and enhanced integration capabilities improved system performance with faster and more accurate data processing and event detection.
A04
Improved testing efficiency
The virtualised and emulated testing environments decrease dependency on real-world vehicle testing, reducing time and resource consumption.
A05
Future-proof architecture
The flexible, device-agnostic platform positioned VisionTrack to easily integrate additional hardware and expand services in the future.
Thanks to this software upgrade, VisionTrack strengthened its position in the sector of AI video telematics thanks to scalable and reliable telemetry software solutions.
The cooperation with Internetum resolved the company’s immediate challenges and laid the groundwork for sustainable growth and ongoing technological development.
