Edge computer
Edge AI unit
AIOT device resides at the network edge that enables AI on IOT devices. Edge AI unit is the foundation of the Atomrock Edge Computing (AEC) architecture and provides connectivity, analytics, management, and many other functions. Different AI algorithms can be used to analyze video and audio stream, wireless packets, and react on IOT devices. By click on the
icon on the right of each device can bring up the device configuration menu.
AEC is the first layer of AI process edge devices to manage video and audio streams. It uses TCP/IP and ONVIF standard to obtain 3rd party video and audio sources and perform AI analysis and cloud recording functions.
- DEVICE / NODE: The device name and the hierarchy it resides in
- STATUS: Online / Offline status
- VIDEO SOURCE: Managed camera video source, click on the name will switch to the camera menu.
- EDIT: The
icon to bring up device configuration menu
Add new edge AI unit
On the upper right of the menu, there are
and
icon. Please use
icon to add new edge AI unit, and
icon to add video source, or delete the edge AI unit.
Add new AEC starts by the
icon on the upper right of the AEC page. Add new AEC by using its serial number. Please connect the AEC to the network and ensure it has AtomCloud connectivity before this step.
- TYPE: The license key type and its description.
- SUBKEY: The license subkey used to enable the function.
- START DATE: The start date of this subkey. This is not changeable once activated.
- EXPIRE DATE: The expire date of this subkey.
After select cloud connection licenses key, chose the node and NTP profile to complete the process.
Click on the
on the right of the device icon will bring up the configuration menu for the edge AI device.
Configuration
- DEVICE NAME: The name of the device, editable via the table
icon. - NODE NAME: The node this device reside in, can use the drop-down list to change it.
- MODEL: The device's model.
- SERIAL NUMBER: The device's serial number.
- STATUS: The device's current online or offline status (cloud connection status).
Licenses
License is required to provide cloud connections or different AI options. Video and audio related license applied on the cameras. Cloud Connection license is required for all units.
If there is multiple of same license type, it will be put together and can be expanded with
icon. The start and end date indicates each subkey's validation date.
ADD:
is used for new added license when there is no same type of license on the device.
EDIT:
is used for extend this type of license with
icon, or delete the license with
icon.
When adding license, please select type then select the subkey used. It will start from today for one year.
When extending license, please select the subkey used. It will extend non-expired the license for another year. If the license is expired, it will start from today for one year.
System
- CONNECTION STATUS: Current connection to cloud status and the total up time. If the device is currently offline, it will display the last connected time.
- CPU USAGE: Current CPU utilization %. The CPU usage might change rapidly, and user can refresh few times to get different readings.
- MEMORY USAGE: Current memory utilization %. The memory usage might change rapidly, and user can refresh few times to get different readings.
- SYSTEM TIME: The time on the device reflect to its time own zone setting.
- REBOOT: Reboot the device, can take several minutes to resume services.
- FIRMWARE VERSION: Version number of the firmware.
- CHECK FOR UPDATE: List if there is a newer version to be updated.
- STREAM PROTOCOL: Select stream protocol for RTSP streaming for all cameras connect to this AEC. Default is TCP. Automatic AEC reboot is required immediately after change.
- NTP SERVER: Set NTP server to enable or disable.
- LOCAL STORAGE: Set Local storage to enable or disable.
- Use external drive as local cache to expand the storage capacity for cloud disconnect scenario. Data in the local cache will automatically sync to cloud after re-connected to cloud. The disk will be encrypted and cannot be read by other devices. Supports SD card, USB-A or USB-C types of serial connection disk drive with at least 32G of capacity. Use of SSD with read/write speed 500MBps or more is suggested.
- FACE AI ACCESS: Set face AI access control mode to enable or disable.
- When enabled, system uses face AI to provide access control. It enables fake check function and minimal face detect area selection function. User must stand in front of the camera that has bigger image (closer) than the minimal face detect area, and system will identify the only the biggest user face at a time. Please use the camera configuration page to set these functions.
NOTE: The Local storage and Face AI access control mode require AEC running firmware 3.6.0+
Select the firmware update version and click on
, then confirm for the update.
Select reboot and confirm for the device reboot process.
Select the RTSP stream protocol for the AEC.
Confirm reboot after change.
Network
NETWORK: Display IPv4 and DNS information. Enable or disable DHCP (reboot required). Please disable DHCP if manual IP and DNS setting is required.
Please ensure the IP information is valid and can access Internet. If the IP information was entered incorrectly or cannot connect to the Internet, user can reset the device to factory default, and it will revert to the default IP setting. Once the device is connected back to the AtomCloud, it will retrieve the profile setting and related configuration automatically.
Time
NTP: Display the NTP Profile used and its information. User can select different NTP profile which defines time zone and NTP server it uses. The NTP profile can be managed in the NTP profile section.
LOG
LOG: The log information related to this device and the event time.
Managed attached AIOT camera device
Select
icon on the upper right to enter edit edge AI unit mode. Here user can change the name of the device and the hierarchy node it resides in. Use the right side
icon to add new video source, or
to remove video source.
Click the
at the far right to delete the edge AI unit. User can select multiple edge AI units at the same time.
Note: all video sources need to be removed before the edge AI unit can be deleted from the system.
Add new video source
Please select “scan” icon to find the ONVIF device on the same network as the edge AI unit. Once the scan completed, it will list the IP address and the status. “test” icon will show on the right if the device is not yet added to the system.
Click on the “test” icon to enter the ONVIF account and password information.
Edge AI unit will start to communicate with the ONVIF device and obtain required information for the next step.
Select the stream used for AI and cloud recording. Please set the codec for the stream to H.264. The stream can be up to 1080p resolution, 10FPS and 4096 bit rate and will be used for both AI and recording.
Note: For AEC running 3.5.2 or lower version needs to select the stream used for AI and recording separately. It can be the same stream ID from camera and it will use two stream traffic.
Please select appropriate cloud connection, cloud recording and AI function licenses. User can only select one license per type. For extending the license, please use the configuration feature under camera menu.
Enter the name of the device to complete the process.
