Edge systems face higher visual complexity than ever. Robots navigate cluttered floors, factory vehicles respond to fast-changing surroundings, and outdoor machines interpret scenes filled with motion, glare, and unpredictable conditions. These shifts demand compute units built to handle dense camera input, richer sensor streams, and AI models running at higher speeds.
Traditional embedded controllers struggle because multi-sensor synchronization becomes harder. Latency pushes safety margins. Modern autonomous systems require platforms that can handle multi-camera setups, offer wide interface flexibility, and support software stacks aligned with today’s AI pipelines. And that’s just the tip of the requirement iceberg!
In this blog, you’ll see how e-con Systems’ Darsi Pro, an AI Vision Box, makes a world of difference for industries looking for next-gen, one-stop-shop vision solutions.
Darsi Pro: e-con Systems’ Answer to a Global Vision Problem
Darsi Pro is an AI Vision Box based on NVIDIA® Jetson Orin™ NX – acting as a single rugged, production-ready compute platform for next-gen mobility systems and other AI vision applications. Powered by NVIDIA® Jetson Orin NX module, it brings advanced AI processing, seamless multi-sensor handling, wide temperature support, and industrial-grade design for real-time perception in dynamic environments.
Currently, Darsi Pro supports a wide range of e-con Systems’ GMSL camera modules, such as:
- STURDeCAM31: 3MP Sony ISX031 120dB HDR Camera
- STURDeCAM88: 4K 140dB HDR Front View Camera
- STURDeCAM57: 5MP RGB-IR Global shutter Camera
View all our compatible GMSL cameras.
Darsi Pro brings advanced AI compute, synchronized sensor integration, and rugged field-ready design, making it ideal for next-generation Physical AI workloads in autonomous systems. Its key capabilities include:
- Darsi Pro uses a fanless build and a rugged enclosure to support consistent performance in locations where dust, vibration, or outdoor exposure is prevalent.
- The AI Vision Box offers support for up to 8 GMSL2 cameras through FAKRA connectors, enabling dense camera layouts for vision-driven applications.
- Interface connectivity options include CAN (x2), PoE(x2), and USB 3.2 Gen 1(x2), RS485, 3 x GPIO, giving developers multiple paths to link sensors, accessories, and supporting equipment.
- NVIDIA ROS2 support helps development teams to work with robotics frameworks that rely on ROS-based workflows.
Why Darsi Pro Is At the Edge of AI Vision Innovation
Delivering multi-sensor fusion
Darsi Pro can work with RADAR, LiDAR, cameras, IMUs, and other sensors through PTP support. It helps developers pull inputs from different sources with predictable timing. The feature supports robotics and mobility use cases that rely on aligned perception signals. The AI Vision Box can handle these inputs in real time.
Enabling cloud management and OTA
Our AI Vision Box is compatible with e-con Systems’ CloVis Central™, which enables remote monitoring, configuration, and health checks for deployed units. OTA handling supports updates and version changes through a secure process. So, it becomes easy to manage devices across fleets with fewer site visits. It also supports long-running applications where consistency matters.
Ensuring wide camera range compatibility
Darsi Pro works with a wide portfolio of e-con Systems’ cameras. It includes options for different resolutions, sensors, and interface formats used in robotics or vision tasks. Developers can choose imaging hardware based on lighting, distance, and environment requirements. The Darsi Pro compatibility range gives room to adapt setups without changing the compute box.
Equipping with industrial-grade design
Darsi Pro is developed for demanding environments where exposure cycles and field conditions can shift quickly. Its construction helps maintain steady processing in such environments. The design supports deployment in areas with dust, water contact, or temperature variation, ensuring predictable operations during long-duty tasks.
Providing flexible connectivity
Dual GbE with PoE, USB3.2, HDMI, CAN, RS485, IMU and wireless modules open up several deployment paths. These options help developers attach sensors, accessories, and networks without additional hardware layers. The AI Vision Box’s connectivity range supports varied layouts, from mobility and automation to vision-focused projects. It also helps when teams manage field installations with mixed interface needs.
Driving powerful edge AI compute
The NVIDIA Ampere GPU architecture with CUDA cores supports high AI throughput at the edge. Models can be run that depend on strong inferencing while handling continuous vision data. The Darsi Pro profile helps with detection, tracking, and similar workloads.
Offering MCU-based smart power management
MCU support enables sequencing for startup and shutdown, along with control for connected peripherals. The feature also helps manage low-power modes during extended use. Developers gain predictable power management when handling various events, such as wake-up or soft reboots.
Applications Where Darsi Pro Can Be a Game-Changer
- Autonomous Mobile Robots (AMRs)
- Delivery Robots
- Warehouse Automation Systems
- Intelligent Transportation Systems (ITS)
Innovation at the Edge: e-con Systems’ AI Vision Box
e-con Systems has been designing, developing, and manufacturing OEM camera solutions since 2003. Backed by over two decades of experience, we have incubated unparalleled capabilities for developing Edge AI Vision Box solutions. Our expertise spans camera selection, module configuration, driver integration, and image processing. It means that customers are guaranteed a smooth journey, from concept to field deployment.
e-con Systems’ dedicated team of engineers also helps meet any demand for tuning or hardware-level adjustments for edge-ready performance.
Find out how exactly your application and use case can harness the power of our AI Vision Box. Please write to camerasolutions@e-consystems.com, and we’d be happy to help you.
FAQs
- What is e-con Systems’ Darsi Pro used for?
e-con Systems’ Darsi Pro is an AI Vision Box based on NVIDIA® Jetson Orin™ NX and designed for mobility systems and other AI vision applications. It supports perception tasks for AMRs, indoor/outdoor robots, smart DVRs, and intelligent video analytics.
- How many cameras can Darsi Pro support?
The AI Vision Box supports up to 8 GMSL2 cameras through FAKRA connectors, enabling dense multi-camera layouts for advanced vision workloads and mobility-focused deployments.
- What camera modules are currently compatible with Darsi Pro?
e-con Systems’ Darsi Pro currently supports a wide range of GMSL cameras in our portfolio. These include STURDeCAM31 (3MP Sony ISX031 120dB HDR Camera), STURDeCAM88 (4K 140dB HDR Front View Camera), and STURDeCAM57 (5MP RGB-IR Global shutter Camera). Explore all our compatible GMSL cameras.
- How does Darsi Pro handle multi-sensor input for robotics?
e-con Systems’ AI Vision Box works with RADAR, LiDAR, cameras, IMUs, and other sensors through PTP support. This helps developers align inputs with predictable timing for robotics and mobility use cases that rely on coordinated perception signals.
- Does Darsi Pro support remote updates and fleet-level device control?
Yes. Darsi Pro integrates with e-con Systems’ CloVis Central™, enabling remote monitoring, configuration, health checks, and OTA updates for deployed units, which simplifies large fleet management.
Prabu is the Chief Technology Officer and Head of Camera Products at e-con Systems, and comes with a rich experience of more than 15 years in the embedded vision space. He brings to the table a deep knowledge in USB cameras, embedded vision cameras, vision algorithms and FPGAs. He has built 50+ camera solutions spanning various domains such as medical, industrial, agriculture, retail, biometrics, and more. He also comes with expertise in device driver development and BSP development. Currently, Prabu’s focus is to build smart camera solutions that power new age AI based applications.