NXS is a plug-and-play sensor board for robots – it handles the wiring and driver setup automatically, so cameras and sensors are streaming usable data in minutes instead of weeks. It connects over GMSL or CAN-FD and works with the MikroElektronika Click ecosystem of sensors, with ROS 2 topics available through the included host bridge.
No reviews yetBe the first to leave a review for NXS by Aliensense
Maker
📌
Hey ProductHunters 👋 We're the team behind NXS, launching today.
We started building this because we were making robots ourselves. And we kept hitting the same wall: before you can write a single line of code, you lose weeks to wiring, driver debugging, and getting parts to talk to each other.
NXS came from solving this for our own builds first. Plug in a camera or a MikroE Click sensor, connect one cable, and the data is ready to use – the driver is generated, no compatibility research.
Here's what it does:
- Setup: plug in a sensor, the AI-agent skill generates the driver, the host tool loads it. You write nothing by hand.
- Connectivity: GMSL2/3 or CAN-FD, one cable carrying power and data back to your compute
- Sensor-side: mikroBUS socket, so you're not locked into one vendor – the MikroE catalog is huge, and most people already have a few of these boards lying around
Output: ROS 2 topics via the NXS host bridge, hardware-timestamped per sample
The result: you go from opening the box to working sensor data in minutes, not weeks.
Happy to answer all your questions about connectivity, ROS 2 integration, or anything else you’d like to know!
Report
Maker
Hey everyone, I work on the firmware side of this. Something worth adding: ROS 2 is one option here, not a requirement. The board speaks Cyphal, and there's a bridge on the host for ROS 2 people. If your stack is something else, or you're on a flight controller, or you just want to read an I²C register map, that all works with no ROS anywhere. We built it that way because we didn't want the protocol choice to be the thing that ruled us out.
If you want to know how any of it works underneath, drivers, transports, timestamps, Cyphal, I'm in the comments today.
Report
Maker
As a proud team member - happy to say that I loved the adventure all the way!
Our goal is to let developers focus on the perception and control theory stuff, not low-level embedded drivers. I love both worlds, but I definitely know which one I'd gladly spend more time with.
Our product is basically a camera and ANY sensor on the same hardware node. One coax back to the compute module, up to 15 m. Samples arrive in SI units with a timestamp synced within sub-millisecond precision. Cyphal/CAN-FD (that we adore!) multi-node for more challenging projects.
Hope you find it useful! If you're doing perception on a Jetson, tell us what you're running!
Report
Hey everyone, also NXS allows you to connect a set of Sony image sensors through GMSL interface and build your own vision system for your robot.
The image sensor and MikroE sensor could be connected to your system through the same GMSL cable simultaneously.
If you don't need image stream you can use CAN or UART to communicate with your set of MikroE sensors. It's up to you.
If you have any question on how to connect sensors or how to use NXS our team will be more than happy to help.
Report
Maker
We've been building robots, just like many of you. Same issues – sensors, wiring, burning time before the runway ends. NXS fixes that – it connects cameras and sensors to your robot without a hassle. And makes it fast. We want you to give it a try. First batch ships now!
Congrats on launching NXS by Aliensense! The idea behind a sensor board that gets your robot running in minutes is compelling, especially because the benefit is immediately clear to the people who need it. Wishing you a strong Product Hunt launch!
Hey everyone, I work on the firmware side of this. Something worth adding: ROS 2 is one option here, not a requirement. The board speaks Cyphal, and there's a bridge on the host for ROS 2 people. If your stack is something else, or you're on a flight controller, or you just want to read an I²C register map, that all works with no ROS anywhere. We built it that way because we didn't want the protocol choice to be the thing that ruled us out.
If you want to know how any of it works underneath, drivers, transports, timestamps, Cyphal, I'm in the comments today.
As a proud team member - happy to say that I loved the adventure all the way!
Our goal is to let developers focus on the perception and control theory stuff, not low-level embedded drivers. I love both worlds, but I definitely know which one I'd gladly spend more time with.
Our product is basically a camera and ANY sensor on the same hardware node. One coax back to the compute module, up to 15 m. Samples arrive in SI units with a timestamp synced within sub-millisecond precision. Cyphal/CAN-FD (that we adore!) multi-node for more challenging projects.
Hope you find it useful! If you're doing perception on a Jetson, tell us what you're running!
Hey everyone, also NXS allows you to connect a set of Sony image sensors through GMSL interface and build your own vision system for your robot.
The image sensor and MikroE sensor could be connected to your system through the same GMSL cable simultaneously.
If you don't need image stream you can use CAN or UART to communicate with your set of MikroE sensors. It's up to you.
If you have any question on how to connect sensors or how to use NXS our team will be more than happy to help.
We've been building robots, just like many of you. Same issues – sensors, wiring, burning time before the runway ends. NXS fixes that – it connects cameras and sensors to your robot without a hassle. And makes it fast. We want you to give it a try. First batch ships now!
Pazi
Congrats on launching NXS by Aliensense! The idea behind a sensor board that gets your robot running in minutes is compelling, especially because the benefit is immediately clear to the people who need it. Wishing you a strong Product Hunt launch!
@zvonimir_sabljic1 thank you!
Orchestra
Wow! That's huge! Good luck with your launch guys! 💪
@sergiepoe Thank you!