Difference between revisions of "MISC-TN-026: Automated test equipment (ATE) monitoring with SBCSPG gateway, ThingsBoard IoT, and Fledge"

From DAVE Developer's Wiki
Jump to: navigation, search
(Introduction)
(22 intermediate revisions by the same user not shown)
Line 23: Line 23:
  
 
==Introduction==
 
==Introduction==
In essence, this Technical Note (TN) is very similar to [[MISC-TN-024: Automated test equipment (ATE) monitoring with SBCSPG gateway and ThingsBoard IoT platform|''MISC-TN-024: Automated test equipment (ATE) monitoring with SBCSPG gateway and ThingsBoard IoT platform'']]. There is a noticeable difference, however, as a different software stack was used on the edge device. In this case, [https://www.lfedge.org/projects/fledge/ Fledge] was utilized.
+
ORCA-TN-001_—_Running_Armbian_Bullseye_(Debian_11)
  
==Testbed==
+
https://mirrors.dotsrc.org/armbian-dl/tritium-h3/archive/Armbian_22.11.1_Tritium-h3_bullseye_current_5.15.80_minimal.img.xz
The edge device is the [[MISC-TN-004:_Running_Debian_(armbian)_on_SBCSPG#The_hardware_platform|SBCSPG gateway]] running Debian Bullseye. To deploy this distribution onto the target, the same approach described [[ORCA-TN-001_—_Running_Armbian_Bullseye_(Debian_11)|here]] was used. Unlike Orca SBC, SBCSPG is based on a 32-bit ARM Cortex-A7 processor, though. As such, a [https://mirrors.dotsrc.org/armbian-dl/tritium-h3/archive/Armbian_22.11.1_Tritium-h3_bullseye_current_5.15.80_minimal.img.xz different image] was used.
 
  
To install Fledge, the <code>apt</code>-based approach described [https://fledge-iot.readthedocs.io/en/latest/quick_start/installing.html here] was followed. It is worth to remember that some dependencies were handled manually. For instance, <code>autoconf</code> and <code>libtool</code> packages were installed manually.
+
LAN .30
  
===Enabling the connection to the PLC===
+
armbian@sbcspg:~$ pip install modbus_cli
As described in [[MISC-TN-024: Automated test equipment (ATE) monitoring with SBCSPG gateway and ThingsBoard IoT platform|''MISC-TN-024'']], the PLC we want to talk to is connected to a different subnet, namely 192.168.30.0/24. For testing the connection to the PLC, after adding the required subnet configuration, the <code>modbus</code> command line tool was used, which is provided by the <code>modbus_cli</code> package. In the following example, the register mapped ad the address 300 is read.
 
  
<pre class="board-terminal">
 
root@sbcspg:/usr/local/fledge/python/fledge/plugins/north# ip addr add 192.168.30.253/24 dev eth0 label eth0:1
 
 
armbian@sbcspg:~/devel/fledge$ pip install modbus_cli
 
Defaulting to user installation because normal site-packages is not writeable
 
Requirement already satisfied: modbus_cli in /home/armbian/.local/lib/python3.9/site-packages (0.1.8)
 
Requirement already satisfied: colorama in /home/armbian/.local/lib/python3.9/site-packages (from modbus_cli) (0.4.6)
 
Requirement already satisfied: umodbus in /home/armbian/.local/lib/python3.9/site-packages (from modbus_cli) (1.0.4)
 
Requirement already satisfied: pyserial~=3.4 in /usr/local/lib/python3.9/dist-packages (from umodbus->modbus_cli) (3.5)
 
 
[notice] A new release of pip is available: 23.0 -> 23.0.1
 
[notice] To update, run: python3 -m pip install --upgrade pip
 
...
 
 
armbian@sbcspg:~$ ./.local/bin/modbus 192.168.30.42:502 300
 
armbian@sbcspg:~$ ./.local/bin/modbus 192.168.30.42:502 300
 
Parsed 0 registers definitions from 1 files
 
Parsed 0 registers definitions from 1 files
300: 32 0x20
+
300: 33 0x21
</pre>
 
 
 
We then configured Fledge in order to read periodically a couple of PLC's registers. Please note that this is a no-code operation as it is carried out with the help of the native web-based Fledge UI. To access PLC exposed data, the Fledge's <code>modbustcp</code> South plugin was used. Please note that also we had to patch it manually to solve a bug that prevented it from working properly. The patch is detailed [https://github.com/fledge-iot/fledge-south-modbustcp/pull/26/files here]. To make the change effective, we disabled and re-enabled the plugin via the web interface.
 
 
 
[[File:SBCSPG-Fledge-2.png|center|thumb|600x600px|Installing the <code>modbustcp</code> South plugin.]]
 
[[File:SBCSPG-Fledge-3.png|center|thumb|600x600px|Configuring the plugin]]
 
[[File:SBCSPG-Fledge-4.png|center|thumb|600x600px|Configuring the plugin.]]
 
[[File:SBCSPG-Fledge-5.png|center|thumb|600x600px|The plugin is enabled.]]
 
[[File:SBCSPG-Fledge-6.png|center|thumb|600x600px|Fledge provides a native web interface to visualize collected data. ]]
 
  
 +
https://github.com/fledge-iot/fledge-south-modbustcp/pull/26/files
  
=== Enabling the connection to ThingsBoard IoT platform ===
+
Riavviare south plugin disable/enable per via grafica.
In order to upload collected data to ThingsBoard platform, a simple North plugin was developed. To this end, the Microsoft Azure plugin was used as reference. The ThingsBoard North plugin is minimalist and '''unfit for a real-world production environment'''. Nevertheless, it was enough to perform the basic tests described in this document. Before testing this new plugin, the <code>mosquitto-clients</code> package was installed as the plugin invokes the <code>mosquitto_pub</code> command line tool.
 
  
<pre class="board-terminal">
 
root@sbcspg:/usr/local/fledge/python/fledge/plugins/north# apt install mosquitto mosquitto-clients
 
Reading package lists... Done
 
Building dependency tree... Done
 
Reading state information... Done
 
The following additional packages will be installed:
 
  libcjson1 libdlt2 libev4 libmosquitto1 libwebsockets16
 
Suggested packages:
 
  apparmor
 
The following NEW packages will be installed:
 
  libcjson1 libdlt2 libev4 libmosquitto1 libwebsockets16 mosquitto mosquitto-clients
 
0 upgraded, 7 newly installed, 0 to remove and 8 not upgraded.
 
Need to get 715 kB of archives.
 
After this operation, 1,492 kB of additional disk space will be used.
 
Do you want to continue? [Y/n]
 
Get:1 http://deb.debian.org/debian bullseye/main armhf libcjson1 armhf 1.7.14-1 [20.5 kB]
 
Get:2 http://deb.debian.org/debian bullseye/main armhf libdlt2 armhf 2.18.6-1+deb11u1 [46.1 kB]
 
Get:3 http://deb.debian.org/debian bullseye/main armhf libev4 armhf 1:4.33-1 [38.6 kB]
 
Get:4 http://deb.debian.org/debian bullseye/main armhf libmosquitto1 armhf 2.0.11-1 [83.5 kB]
 
Get:5 http://deb.debian.org/debian bullseye/main armhf libwebsockets16 armhf 4.0.20-2 [163 kB]
 
Get:6 http://deb.debian.org/debian bullseye/main armhf mosquitto armhf 2.0.11-1 [253 kB]
 
Get:7 http://deb.debian.org/debian bullseye/main armhf mosquitto-clients armhf 2.0.11-1 [110 kB]
 
Fetched 715 kB in 0s (1,817 kB/s)     
 
Selecting previously unselected package libcjson1:armhf.
 
(Reading database ... 72206 files and directories currently installed.)
 
Preparing to unpack .../0-libcjson1_1.7.14-1_armhf.deb ...
 
Unpacking libcjson1:armhf (1.7.14-1) ...
 
Selecting previously unselected package libdlt2:armhf.
 
Preparing to unpack .../1-libdlt2_2.18.6-1+deb11u1_armhf.deb ...
 
Unpacking libdlt2:armhf (2.18.6-1+deb11u1) ...
 
Selecting previously unselected package libev4:armhf.
 
Preparing to unpack .../2-libev4_1%3a4.33-1_armhf.deb ...
 
Unpacking libev4:armhf (1:4.33-1) ...
 
Selecting previously unselected package libmosquitto1:armhf.
 
Preparing to unpack .../3-libmosquitto1_2.0.11-1_armhf.deb ...
 
Unpacking libmosquitto1:armhf (2.0.11-1) ...
 
Selecting previously unselected package libwebsockets16:armhf.
 
Preparing to unpack .../4-libwebsockets16_4.0.20-2_armhf.deb ...
 
Unpacking libwebsockets16:armhf (4.0.20-2) ...
 
Selecting previously unselected package mosquitto.
 
Preparing to unpack .../5-mosquitto_2.0.11-1_armhf.deb ...
 
Unpacking mosquitto (2.0.11-1) ...
 
Selecting previously unselected package mosquitto-clients.
 
Preparing to unpack .../6-mosquitto-clients_2.0.11-1_armhf.deb ...
 
Unpacking mosquitto-clients (2.0.11-1) ...
 
Setting up libmosquitto1:armhf (2.0.11-1) ...
 
Setting up libev4:armhf (1:4.33-1) ...
 
Setting up libcjson1:armhf (1.7.14-1) ...
 
Setting up mosquitto-clients (2.0.11-1) ...
 
Setting up libdlt2:armhf (2.18.6-1+deb11u1) ...
 
Setting up libwebsockets16:armhf (4.0.20-2) ...
 
Setting up mosquitto (2.0.11-1) ...
 
Created symlink /etc/systemd/system/multi-user.target.wants/mosquitto.service → /lib/systemd/system/mosquitto.service.
 
Processing triggers for libc-bin (2.31-13+deb11u5) ...
 
</pre>
 
  
The following box shows the code of the ThingsBoard plugin.
+
You can think about this This Technical Note (TN) as an evolution of the article "[[MISC-TN-021: SBCSPG and ThingsBoard IoT]]". The use case described here is a real-world application based on the same elements:
 +
* [[MISC-TN-004: Running Debian (armbian) on SBCSPG#The hardware platform|The ''SBCSPG'' industrial gateway]]
 +
* [https://thingsboard.io/docs/ ''ThingsBoard IoT platform'']
 +
* [https://thingsboard.io/docs/iot-gateway/what-is-iot-gateway/ ''ThingsBoard IoT gateway''].
 +
The application is to monitor robotized DAVE Embedded Systems' ATEs like the one shown [https://youtu.be/OoNiKgJXq00?t=409 here].
  
<syntaxhighlight lang="python" class"mw-collapsible="" mw-collapsed"="">
+
== Implementation ==
# -*- coding: utf-8 -*-
+
The implementation is suitable for a production environment. Unlike what we did for [[MISC-TN-021: SBCSPG and ThingsBoard IoT|this TN]], in this case the ThingsBoard IoT platform was installed on the cloud, specifically on an AWS EC2 instance running Ubuntu Linux. The following diagram shows the complete solution.
  
""" ThingsBoard North plugin"""
+
[[File:ATE monitoring SBCSPG ThingsBoard.png|center|thumb|1000x1000px]]
import asyncio
 
import time
 
import json
 
import sys
 
import os
 
  
 +
Technically, the gateway can communicate with the ATE using different physical links and protocols. In the example under discussion, the physical link is Ethernet and the protocol is Modbus/TCP. The gateway retrieves periodically some data regarding the robotic arm used to move the device under test (DUT). These data are [https://www.universal-robots.com/articles/ur/interface-communication/modbus-server/ exposed by the PLC governing the arm] and are documented in detail [https://s3-eu-west-1.amazonaws.com/ur-support-site/16377/ModBus%20server%20data.pdf here].
  
 +
=== ThingsBoard IoT configuration ===
 +
To install ThingsBoard IoT platform on an EC2 instance, [https://thingsboard.io/docs/user-guide/install/cluster/aws-self-hosted-setup/ these instructions] were followed.
  
from fledge.common import logger
+
=== ThingsBoard IoT gateway configuration ===
from fledge.plugins.north.common.common import *
+
To run this esample, the most important module to configure was the [https://thingsboard.io/docs/iot-gateway/config/modbus/ Modbus connector]. The following box shows the resulting configuration file needed to poll some variables of the arm's joints (namely temperatures and currents).<syntaxhighlight lang="json">
 
+
armbian@sbcspg:~$ cat /etc/thingsboard-gateway/config/modbus.json
__author__ = "Andrea Marson"
+
{
__copyright__ = "Copyright (c) 2023 DAVE Srl"
+
    "master": {
__license__ = "Apache 2.0"
+
      "slaves": [
__version__ = "${VERSION}"
+
        {
 
+
          "host": "192.168.30.42",
_LOGGER = logger.setup(__name__)
+
          "port": 502,
 
+
          "type": "tcp",
_CONFIG_CATEGORY_NAME = "THINGSBOARD"
+
          "method": "socket",
_CONFIG_CATEGORY_DESCRIPTION = "ThingsBoard Python North Plugin"
+
          "timeout": 35,
 
+
          "byteOrder": "BIG",
_DEFAULT_CONFIG = {
+
          "retries": true,
    'plugin': {
+
          "retryOnEmpty": true,
        'description': 'ThingsBoard North Plugin',
+
          "retryOnInvalid": true,
        'type': 'string',
+
          "pollPeriod": 5000,
        'default': 'thingsboard',
+
          "unitId": 1,
        'readonly': 'true'
+
          "deviceName": "UR5",
    },
+
          "sendDataOnlyOnChange": true,
    "tb_iot_platform_host": {
+
          "connectAttemptTimeMs": 5000,
            "description": "ThingsBoard IoT platform host.",
+
          "connectAttemptCount": 5,
            "type": "string",
+
          "waitAfterFailedAttemptsMs": 300000,
            "default": "put the public name of your host here",
+
          "timeseries": [
             "order": "2",
+
            {
            "displayName": "ThingsBoard IoT platform host"
+
              "tag": "base_joint_temperature_deg_C",
    },
+
              "type": "16int",
    "tb_mqtt_topic": {
+
              "functionCode": 3,
            "description": "MQTT topic",
+
              "objectsCount": 1,
            "type": "string",
+
              "address": 300
            "default": "v1/devices/me/telemetry",
+
            },
             "order": "3",
+
             {
            "displayName": "MQTT topic",
+
              "tag": "shoulder_joint_temperature_deg_C",
            'readonly': 'true'
+
              "type": "16int",
    },
+
              "functionCode": 3,
    "tb_device_access_token": {
+
              "objectsCount": 1,
            "description": "Device's access token.",
+
              "address": 301
            "type": "string",
+
            },
            "default": "put your device's token here",
+
            {
             "order": "4",
+
                "tag": "base_joint_current_mA",
            "displayName": "Device's access token"
+
                "type": "16int",
    },
+
                "functionCode": 3,
    "source": {
+
                "objectsCount": 1,
            "description": "Source of data to be sent on the stream.",
+
                "address": 290
            "type": "enumeration",
+
            },
            "default": "readings",
+
             {
             "options": ["readings"],
+
                "tag": "shoulder_joint_current_mA",
            "order": "5",
+
                "type": "16int",
            "displayName": "Source",
+
                "functionCode": 3,
             'readonly': 'true'
+
                "objectsCount": 1,
 +
                "address": 291
 +
            },
 +
            {
 +
                "tag": "elbow_joint_current_mA",
 +
                "type": "16int",
 +
                "functionCode": 3,
 +
                "objectsCount": 1,
 +
                "address": 292
 +
            },
 +
             {
 +
              "tag": "wrist1_joint_current_mA",
 +
              "type": "16int",
 +
              "functionCode": 3,
 +
              "objectsCount": 1,
 +
              "address": 293
 +
            },
 +
            {
 +
                "tag": "wrist2_joint_current_mA",
 +
                "type": "16int",
 +
                "functionCode": 3,
 +
                "objectsCount": 1,
 +
                "address": 294
 +
            },
 +
             {
 +
                "tag": "wrist3_joint_current_mA",
 +
                "type": "16int",
 +
                "functionCode": 3,
 +
                "objectsCount": 1,
 +
                "address": 295
 +
             }
 +
          ]
 +
        }
 +
      ]
 
     }
 
     }
}
+
  }
 +
</syntaxhighlight>
  
def plugin_info():
+
== Dashboards ==
    return {
+
ThingsBoard IoT allows to create easily custom dashboards to visualize data. At [http://ec2-15-160-47-190.eu-south-1.compute.amazonaws.com:8080/dashboard/1178fb20-e56d-11ec-8592-e3cb29424137?publicId=28967df0-e56d-11ec-8592-e3cb29424137 this link] such a dashboard is available showing some temperatures and currents of the arm's joints.
        'name': 'thingsboard',
+
[[File:ATE monitoring SBCSPG ThingsBoard dashboard1.png|center|thumb|600x600px]]
        'version': '2.1.0',
 
        'type': 'north',
 
        'interface': '1.0',
 
        'config': _DEFAULT_CONFIG
 
    }
 
 
 
def plugin_init(data):
 
    _LOGGER.info('Initializing ThingsBoard North Python Plugin')
 
    global thingsboard_north, config
 
    thingsboard_north = ThingsBoardNorthPlugin()
 
    config = data
 
    _LOGGER.info(f'config = {config}')
 
    return config
 
 
 
"""
 
Example of payload
 
 
 
[{'reading': {'base_joint_temperature_deg_C': 34, 'base_joint_current_mA': 2042}, 'asset_code': 'Modbus TCP', 'id': 35613, 'ts': '2023-02-22 16:33:49.961000+00:00', 'user_ts': '2023-02-22 16:33:47.126849+00:00'}, {'reading': {'base_joint_temperature_deg_C': 34, 'base_joint_current_mA': 2045}, 'asset_code': 'Modbus TCP', 'id': 35614, 'ts': '2023-02-22 16:33:49.961000+00:00', 'user_ts': '2023-02-22 16:33:48.126817+00:00'}, {'reading': {'base_joint_temperature_deg_C': 34, 'base_joint_current_mA': 2060}, 'asset_code': 'Modbus TCP', 'id': 35615, 'ts': '2023-02-22 16:33:49.961000+00:00', 'user_ts': '2023-02-22 16:33:49.126827+00:00'}]
 
 
 
Example of upload
 
 
 
mosquitto_pub -d -q 1 -h "$THINGSBOARD_HOST_NAME" -t "v1/devices/me/telemetry" -u "$ACCESS_TOKEN" -m '{"base_joint_temperature_deg_C":33}'
 
"""
 
async def plugin_send(data, payload, stream_id):
 
    try:
 
        _LOGGER.info(f'ThingsBoard North Python - plugin_send: {stream_id}')
 
        _LOGGER.info(f'data = {data}')
 
        _LOGGER.info(f'payload = {payload}')
 
        _LOGGER.info(f'host = {data["tb_iot_platform_host"]["value"]}')
 
        for reading in payload:
 
            _LOGGER.info(f'reading = {reading}')
 
            _LOGGER.info(f'reading[reading] = {reading["reading"]}')
 
            s = 'mosquitto_pub -d -q 1 -h "' + data["tb_iot_platform_host"]["value"] + '" -t "' + data["tb_mqtt_topic"]["value"] + '" -u "' + data["tb_device_access_token"]["value"] + '" -m \'' + str(reading["reading"]) + "\'"
 
            _LOGGER.info(f's = {s}')
 
            os.system(s)
 
    except asyncio.CancelledError as ex:
 
        _LOGGER.exception(f'Exception occurred in plugin_send: {ex}')
 
    else:
 
        _LOGGER.info('payload sent successfully')
 
        return True, payload[len(payload-1)]["id"], len(payload)
 
 
 
def plugin_shutdown(data):
 
    pass
 
 
 
# TODO: North plugin can not be reconfigured? (per callback mechanism)
 
def plugin_reconfigure():
 
    pass
 
 
 
class ThingsBoardNorthPlugin(object):
 
    """ North ThingsBoard Plugin """
 
 
 
 
 
</syntaxhighlight>
 
== Visualization of uploaded data ==
 
Similarly to what we did in [[MISC-TN-024: Automated test equipment (ATE) monitoring with SBCSPG gateway and ThingsBoard IoT platform|''MISC-TN-024'']], we created a simple dashboard to visualize in real-time on cloud side collected data as shown in the following example.
 
[[File:SBCSPG-Fledge-8.png|center|thumb|600x600px|Visualization of uploaded data.]]
 

Revision as of 12:00, 20 February 2023

Info Box
SBCSPG-1.jpg Applies to SBCSPG
Cloud-computing-banner.jpg Applies to IoT
Warning-icon.png This technical note was validated against specific versions of hardware and software. It may not work with other versions. Warning-icon.png


History[edit | edit source]

Version Date Notes
1.0.0 February 2023 First public release

Introduction[edit | edit source]

ORCA-TN-001_—_Running_Armbian_Bullseye_(Debian_11)

https://mirrors.dotsrc.org/armbian-dl/tritium-h3/archive/Armbian_22.11.1_Tritium-h3_bullseye_current_5.15.80_minimal.img.xz

LAN .30

armbian@sbcspg:~$ pip install modbus_cli

armbian@sbcspg:~$ ./.local/bin/modbus 192.168.30.42:502 300 Parsed 0 registers definitions from 1 files 300: 33 0x21

https://github.com/fledge-iot/fledge-south-modbustcp/pull/26/files

Riavviare south plugin disable/enable per via grafica.


You can think about this This Technical Note (TN) as an evolution of the article "MISC-TN-021: SBCSPG and ThingsBoard IoT". The use case described here is a real-world application based on the same elements:

The application is to monitor robotized DAVE Embedded Systems' ATEs like the one shown here.

Implementation[edit | edit source]

The implementation is suitable for a production environment. Unlike what we did for this TN, in this case the ThingsBoard IoT platform was installed on the cloud, specifically on an AWS EC2 instance running Ubuntu Linux. The following diagram shows the complete solution.

ATE monitoring SBCSPG ThingsBoard.png

Technically, the gateway can communicate with the ATE using different physical links and protocols. In the example under discussion, the physical link is Ethernet and the protocol is Modbus/TCP. The gateway retrieves periodically some data regarding the robotic arm used to move the device under test (DUT). These data are exposed by the PLC governing the arm and are documented in detail here.

ThingsBoard IoT configuration[edit | edit source]

To install ThingsBoard IoT platform on an EC2 instance, these instructions were followed.

ThingsBoard IoT gateway configuration[edit | edit source]

To run this esample, the most important module to configure was the Modbus connector. The following box shows the resulting configuration file needed to poll some variables of the arm's joints (namely temperatures and currents).

armbian@sbcspg:~$ cat /etc/thingsboard-gateway/config/modbus.json
{
    "master": {
      "slaves": [
        {
          "host": "192.168.30.42",
          "port": 502,
          "type": "tcp",
          "method": "socket",
          "timeout": 35,
          "byteOrder": "BIG",
          "retries": true,
          "retryOnEmpty": true,
          "retryOnInvalid": true,
          "pollPeriod": 5000,
          "unitId": 1,
          "deviceName": "UR5",
          "sendDataOnlyOnChange": true,
          "connectAttemptTimeMs": 5000,
          "connectAttemptCount": 5,
          "waitAfterFailedAttemptsMs": 300000,
          "timeseries": [
            {
              "tag": "base_joint_temperature_deg_C",
              "type": "16int",
              "functionCode": 3,
              "objectsCount": 1,
              "address": 300
            },
            {
              "tag": "shoulder_joint_temperature_deg_C",
              "type": "16int",
              "functionCode": 3,
              "objectsCount": 1,
              "address": 301
            },
            {
                "tag": "base_joint_current_mA",
                "type": "16int",
                "functionCode": 3,
                "objectsCount": 1,
                "address": 290
            },
            {
                "tag": "shoulder_joint_current_mA",
                "type": "16int",
                "functionCode": 3,
                "objectsCount": 1,
                "address": 291
            },
            {
                "tag": "elbow_joint_current_mA",
                "type": "16int",
                "functionCode": 3,
                "objectsCount": 1,
                "address": 292
            },
            {
              "tag": "wrist1_joint_current_mA",
              "type": "16int",
              "functionCode": 3,
              "objectsCount": 1,
              "address": 293
            },
            {
                "tag": "wrist2_joint_current_mA",
                "type": "16int",
                "functionCode": 3,
                "objectsCount": 1,
                "address": 294
            },
            {
                "tag": "wrist3_joint_current_mA",
                "type": "16int",
                "functionCode": 3,
                "objectsCount": 1,
                "address": 295
            }
          ]
        }
      ]
    }
  }

Dashboards[edit | edit source]

ThingsBoard IoT allows to create easily custom dashboards to visualize data. At this link such a dashboard is available showing some temperatures and currents of the arm's joints.