Application Case | JHCTECH AI Edge Computing Empowers Low-speed Unmanned Vehicles

  • Home
  • Media
  • News
  • Application Case | JHCTECH AI Edge Computing Empowers Low-speed Unmanned Vehicles

Application Case | JHCTECH AI Edge Computing Empowers Low-speed Unmanned Vehicles

Preface:

Low-speed unmanned vehicles usually refer to autonomous vehicles operating in relatively simple and fixed scenarios at low speeds, also known as "low-speed autonomous driving systems." As a comprehensive carrier of cutting-edge technologies such as intelligence, electrification, and informatization, low-speed unmanned vehicles are an important part of future intelligent transportation and smart city development. They serve as the foundation for the evolution of the next generation of intelligent ground transportation systems and are of great significance to the advancement of intelligent transportation.

 

Application Requirements for Low-Speed Unmanned Vehicles

The core technology of autonomous driving generally consists of three levels: perception, decision-making, and execution. Among these, the decision-making layer, considered the “brain” of autonomous driving, is crucial. A low-speed unmanned vehicles supplier found JHCTECH to work together to develop a high-quality low-speed unmanned vehicle kit. As the support for the unmanned vehicles decision-making layer, the on-board edge computing device is responsible for collecting information from various environmental sensors such as lidar and cameras, and virtually reconstructing road conditions. Simultaneously, it determines parameters such as driving path and speed based on various information like maps, playing a key role in the operation of the unmanned vehicle.

Vehicle-mounted MEC equipment is used in mobile robots and low-speed unmanned vehicles with L4 autonomous driving capabilities, where stability and reliability are top priorities. These products must meet several stringent requirements, including CPU computing power, storage capacity, data security, and equipment safety.

  • High-performance processor, supporting AI/GPU cards for fast processing of data from vehicle-mounted perception units
  • Multiple storage and high-speed storage capabilities
  • Rich IO interfaces to meet the connection of multiple peripheral devices
  • CAN bus to interface with walking/steering actuators
  • Wide voltage DC power supply, suitable for vehicle-mounted battery power
  • Anti-vibration design, adaptable to vehicle-mounted environments

 

JHCTECH’s In-Vehicle Solutions

Low-speed unmanned vehicles rely on on-board sensors to perceive their surroundings and rapidly process data from perception units to ensure safe autonomous driving. JHCTECH's BRAV-7520-WP connects on-board laser radar, obstacle avoidance radar, gyroscope, sensor, face recognition, etc. to perform deep learning reasoning calculations and data structured fusion.

Adopting the dual architecture solution of CPU+GPU, the device is equipped with Intel® Xeon® E or 9th/8th-Gen Core™ processor. The WP model features 1*PCIe X16 (X16 signal) and supports up to 350W graphics card. In this project, the RTX-3080 high-computing power GPU card is used for deep learning to achieve the structured data of radar and vision fusion. The DC 9-55V wide-voltage power supply is compatible with vehicle battery systems. Equipped with rich IO interfaces, it can meet the connection of multiple peripheral devices such as laser radar, millimeter wave radar, camera, etc.; It supports ultra-high-definition dual 4K and three independent displays (2DP, 1*VGA), connected to vehicle displays. The chassis wire control system is connected via the CAN bus to achieve braking and steering of low-speed unmanned vehicles. The overall structure and installation method of BRAV-7520-WP are designed according to the shock absorption scheme, suitable for the vehicle environment and ensuring robustness and stability.

 

BRAV-7520-WP

  • Intel® Xeon®E or 9th/8th Gen Core™ CPU
  • Ultra HD dual 4K, three independent displays 2*DP, 1*VGA
  • 3*Gig-LAN (iATM supported), optional multi-channel 10G optical port card
  • Dual PCIe standard slots, supporting various high-speed expansion function cards
  • Multiple storage options: 2*SATA3.0, 1*M.2 M-Key, support NVMe
  • Fanless CPU, efficient air cooling design for AI/GPU cards
  • Maximum total output power of 600W, supports either one 350W GPU card or two 75W AI accelerator cards
  • Wide voltage DC power supply DC 9-55V, with over-voltage, over-current, and reverse polarity protection

 

JHCTECH ARM-Based In-Vehicle Solution Recommendation

In addition to the Intel X86 architecture solutions, JHCTECH also offers ARM-based solutions featuring NVIDIA Jetson NX Orin and Jetson AGX Orin with the BRAV-7121 and BRAV-7134. Let’s take a closer look together!

BRAV-7121

  • NVIDIA Jetson Orin NX 8/16G,70/100TOPS
  • 6/8*Cores ARM Cortex-A78AE v8.2 64 bit CPU
  • Ampere GPU(1024 cores), 32*Tensor cores
  • 8/16G LPDDR5, 1*M.2 2280 M-Key NVMe
  • 2*LAN(POE optional), 2*Iso.CAN (One of the CAN channels with isolation is optional), 4*GMSL(optional);
  • Video code:1x4k60 | 3x4k30 | 12x1080p30 H265
  • Video decode:1x8k30 | 4x4k30 | 18x1080p30 H265
  • 1*HDMI,1*Line-out ,2*USB3.2, 2*USB2.0, 1*Debug, 2*Iso.COM,1*8-bit Iso.DIO
  • 1*M.2 3052 B-Key+SIM; 1*MiniPCIe;
  • DC-IN 9~36V wide power input, DC-OUT 12V;
  • Aluminum alloy chassis, fanless cooling design

 

BRAV-7134

  • NVIDIA Jetson AGX Orin 32/64G, 200/275TOPS
  • 8/12*Cores ARM Cortex-A78AE CPU, 2.2GHz (maximum frequency)
  • NVIDIA Ampere GPU with Tensor Cores
  • Onboard 32/64G 256-bit LPDDR5 RAM and 64GB eMMC, 2*M.2 M-Key NVME
  • 1*HDMI, 1*Audio Line-out, 2*LAN, 6 Gigabit Ethernet ports on switches, 2*USB3.2,2*USB2.0, 1*USB2.0 (inside), 2*COM, 8-bit DIO
  • DC9~36V wide power input, with OVP, OCP, and SCP
  • Aluminum-magnesium alloy chassis, Fan cooling design
  • Maximum support -30-80°C temperature range
 2024-10-16
0