Overview
The inception of Artificial Intelligence (AI)technology is shifting the paradigm of Industry 4.0 from Automation to Autonomy, expediting the development of Autonomous Mobile Robots (AMRs) and Autonomous Vehicles (AVs). One of the results is that the automated transporters used in factories, mines, warehouses, construction sites and many other industrial and commercial domains no longer move on guided rails; instead, with GPS mapping, advanced sensors and AI algorithms, they freely move on floors, acting on data and learning from data, transporting goods to destinations without creating collisions with people and other autonomous vehicles..
A California-based developer for autonomous driving software solutions has taken a step further by creating a solution that not only can be swiftly integrated onto newly manufactured vehicles but also can be applied to retrofits of many types of existing industrial vehicles owned by their customers. A flexible and scalable solution that has never been seen before, it eliminates traditional barriers to autonomous vehicle adoption, such as high costs and complex installations, and turns existing vehicles into autonomous fleets running at the forefront of Industry 4.0 with minimum investment and effort.
The implementation of their ground-breaking and game-changing technology relies on a robust edge AI computer with compelling computing power, outstanding graphic processing, and uncompromised ruggedness. This Californian self-driving developer had tried a few brands of computers in their previous efforts but failed to resolve shock and vibration issues, and the add-on graphics card would sometimes be shook out of the slot. Eventually, they found Neousys’ Nuvo-8108GC, a rugged edge AI GPU computer that matches their technology and helps to bring autonomy to their customers’ existing industrial vehicles and realize benefits of easy deployment, reduced labor intensity, lowered costs, improved safety, scaled-up efficiency, and data insights that can improve operations.
Challenges
Extra ruggedness for mobile applications
Autonomous vehicles bring computers to the edge and they must endure constant shocks and vibrations. In many industrial settings, adverse conditions could also involve variable temperature, poor ventilation and high humidity. Computers used for industrial autonomous vehicles, therefore, must be rugged and robust enough to endure the complex and harsh environments for reliable day-to-day operations.
Graphics card processing for AI
The AI-based autonomous vehicle technology developed by this Californian company can detect more than 1,000 objects per second, implying the need for a high performance edge AI computer with powerful GPU support to process real time images and sensor data acquired from the surrounding environment of the moving vehicle.
Ignition power control
During a car's start-up/ shutdown, voltage spikes could damage the in-vehicle computer, which can be avoided if the computer is provisioned with a carefully designed ignition control function that can delay the computer from powering up/shutting down and shun the voltage spikes.
Flexibility and expandability
Being a single unified solution customizable for various types of vehicles in different settings, the AV kit developed by this Californian company needs an in-vehicle computer providing rich and flexible I/O to support different needs and to expand functions for today and tomorrow.
Solution
The Californian company provides turnkey self-driving solution for both retrofit and new vehicles, such as haulers, stockchasers, tow tractors and other utility vehicles. Their solution unifies robotaxi-level autonomous driving software, advanced sensing units, precise analytics and best-of-the-class hardware—including sensing units, a human-machine interface, a lighting system, and a leading industrial computer like Neousys Nuvo-8108GC sitting at the heart of the system.
Advantages of Neousys Edge AI GPU Computers
- Featuring the latest NVIDIA® RTX 30 series graphics card secured with a Neousys-patented bracket to withstand vibrations, as well as a thermal design for -25°C to 60°C operations, it is one of the best edge AI GPU computers available off-the-shelf for rugged mobile applications.
- Built-in ignition control and 8~48V wide-range DC input for protecting the computer from voltage spikes during car start-ups/shut-downs.
- Rich and flexible I/O provides high expandability for customization, wireless communication, and future-proof designs, including 2x PCIe x16 slot@Gen3 of 8-lanes, 2x PCIe x8 slots@Gen3 of 4-lanes, 2x M.2 B key and 2x full-size mini-PCIe sockets.
Benefits of Neousys Rugged Edge AI Computers
Neousys rugged edge AI computers offer the following advantages:
Environmental
- Rugged computers for extra volatile environmental conditions
- True -25°C up to 60°C wide-temperature operations for harsh, volatile environments
- MIL-STD-810G certified
- Tri-axis tested patented damping bracket to withstand shock and vibration conditions
- Dedicated bracket to secure graphics card in-place in in-vehicle conditions
Connectivity/ expandability
- USB3.1 Gen1/ Gen2 connectivity for USB cameras
- Connection ports with screw-lock for rugged connectivity
- WiFi 6/ WiFi 5/ 5G/ 4G wireless communication, expansion via M.2/ mPCIe module
Inference processing power
- Support NVIDIA® Tesla/ Quadro inference accelerators
- Support NVIDIA® RTX 30/ 20 series graphics cards
Electrical
- Patented SuperCAP UPS to counteract unforeseen power interruptions
- Wide-range DC input
Expansion
- PCIe expansion for 10G/ 5G/ 2.5G/ PoE cards
- PCI expansion for legacy connectivity
- mini-PCIe expansion for WiFi wireless modules
- M.2 B/ E key expansion for 5G/ 4G modules
- M.2 M key expansion for NVMe/ SATA signal SSD