IoT
4 min read
Oct 26, 2022
We have already discussed the chipsets we worked with in one of our projects with static IoT devices. Now time is coming to know more about chipsets for moving IoT devices. So, NVIDIA chipsets why the Customer gave his heart to it.
NVIDIA boards became famous and got a reputation among gamers and graphics designers (GeForce series) a time ago, and now NVIDIA has Jetson series.
The first board was the TX1 released in November, 2015. Now NVIDIA has released the more powerful and power-efficient Jetson TX2 board.
The Jetson boards are siblings to NVIDIA’s Drive PX boards for autonomous driving and the TX2 shares the same Tegra “Parker” silicon as the Drive PX2.
There are many synergies between the two families as both can be used to add local machine learning to transportation. The Drive PX boards are designed for automotive with extended temperature ranges and high reliability requirements. The Jetson boards are optimized for compact enclosures and battery power for smaller, portable equipment.
With devices such as robots, drones, 360 cameras, medical, etc., Jetson can be used for “edge” machine learning. The ability to process data locally and with limited power is useful when connectivity bandwidth is limited or spotty (like in remote locations), latency is critical (real-time control), or where privacy and security is a concern.
Another innovative solution from NVIDIA - Jetson Nano.
Jetson Nano development board is also a powerful small artificial smart computer, which only needs to insert a MicroSD card with a system image to start, built-in SOC system-level chip, can have a parallel hand, such as Tensorflow, Pytorch, Caffe / Caffe2, Keras, MXNET and other neural networks that can be used to achieve functionality such as image class, target detection, speech segmentation, and intelligent analysis. Usually used to build autonomous robots and complex artificial intelligence systems.
The Customer had chosen this chip for his moving device, because it was extremely important to detect obstacles and define direction. All the tasks were covered by the chipset functionality rather successfully.
You may ask why not to choose Raspberry Pi all the more reasonably priced by the way.
Raspberry was considered as an alternative. In fact, they are actually very similar in primary functions, and all can develop some special functions, such as ARM processors, 4GB RAM, and a series of peripherals.
As for video-out: the Nano has both HDMI 2.0 and DisplayPort available, which can be used at the same time. The Pi is limited to either its HDMI port or its proprietary display interface, which as far as we at Inmost know cannot be used simultaneously.
They both have multiple ways of interfacing, including I2C, I2S, serial, and GPIO, but we also appreciate that the Nano has USB3.0 and Gigabit Ethernet.
However the biggest difference is that the Raspberry Pi has a low power VideoCore multimedia processor, and Jetson Nona contains higher performance, more powerful GPUs (graphics processors), which makes it support some functions that Raspberry Pi Can't do. Then Jeston Nona makes some more depth developments possible and has more potential in development.
For our customer's project, fast processing of video from the camera is the number one task, so it was clearly decided to use Jetson Nano to solve this problem.
The NVIDIA Jetson system is high performance and power-efficient, making it one of the best and most popular platforms for building machines based on AI on the edge (Edge Machine Learning).
