Hardware Requirements
LPU
LPU, LiDAR Processing Unit, refers to the computing device that SENSR-I is running on. Seoul Robotics selects and sells 4 types of edge computing devices as Standard LPU, which is LPU A~D. If you buy Seoul Robotics' U.S.S. Discovery or Endeavor, SENSR-I software is provided as pre-installed on the LPU.
Each LPU can be selected according to the following criteria.
| LPU Types | Max # of connectable LiDARs | GPU module availability |
|---|---|---|
| LPU A | 1 | X |
| LPU B | 2 | O |
| LPU C | 8 | X |
| LPU D | 8 | O |
The maximum number of LiDARs that can be connected can vary depending on the area you want to view using LiDAR. Please get consulted by the Sales Representative at Seoul Robotics.
Requirement for your own computing device
If you intend to install the software on your own provided hardware, a couple of reference examples of minimum hardware requirements are as follows:
Single Sensor, producing < 400,000 points per second
- Intel Celeron J1900, Quad-core, 2.0GHz
- 8GB Memory
- 64GB HDD (preferably solid state)
- 1+ x 1Gbps Ethernet
- Ubuntu 20.04 LTS
- Does NOT support SENSR2.DL GPU module!
Single or multiple sensors, producing < 3,000,000 points per second total
- NVidia Xavier NX, 6-core Carmel, ARM v8.2
- 8GB Memory
- 128GB M.2 2280 SSD
- Jetpack 5.0 and above
- Supports SENSR2.DL GPU module!
Up to 8 sensors, each producing < 350,000 points per second, or 6,000,000 total
- Intel 8th+ Gen i5 (i7 preferred) processor, >2.0Ghz
- 8GB Memory
- 64GB HDD (preferably solid state)
- 2x 1Gbps Ethernet (6x preferred)
- Ubuntu 20.04 LTS
- (Optional) NVidia GTX1650/Quadro RTX5000 or above to support SENSR2.DL GPU Module.
Above 8 sensors
Please contact Seoul Robotics for recommendations.
The exact hardware requirements for optimal performance are dependent on the number of connected sensors, the amount of data produced by each sensor, the total area coverage and the maximum number of simultaneously tracked objects in the scene. The above are our minimum recommendations.
Requirements for SENSR2.DL GPU Module
SENSR 2 introduces an optional module that contains a number of advanced Deep Learning based algorithms that will improve the tracking and classification performance significantly.
The requirements to take advantage of this technology are one of the following:
- NVidia Xavier
- NVidia GTX 1650 (minimum) or 2080 (recommended) or above.
- NVidia RTX4000
This module is activated by your license. See Project Setup for how to enable the module and Runtime mode for more details.
Operating System
SENSR will only run on the following Operating Systems
- Ubuntu 20.04 LTS
- Jetpack 5.0 or above (for ARM devices)
If you need to run SENSR on a different OS, please contact Seoul Robotics.
Other Requirements
- SENSR has a GUI that uses OpenGL. The computer must have an OpenGL compatible graphics card.
- Make sure the /opt/ directory is read/write enabled. SENSR installs the project files and environment settings by default in the /opt/seoulrobotics folder.