Dashboard Research - Niklas#2


Aim of this post

I am going to share my experiences of this project during the last two weeks. It were two hard weeks with moments of happiness and frustration. At the end, you will find my recommendations for the continuations of the project. Let's start. 


Description of task and issue:

Task

The main task for the next weeks was to get an IOT Platform with integrated widgets. It should be able to do the following:
  • be connectable to a MQTT-broker
  • support micro:bit
  • support the esp8266 Wi-Fi module
  • support using widgets: 
    • a map (using GPS)
    • setting waypoints on this map
    • a camera feed (like a live-video as widget)
    • ability to control and view digital/analog input/output
  • be free to use, or at least affordable for a University
The dashboard itself should be creatable with MakeCode.



Gamechanger

We already had a micro:bit with an esp8266 connected to an IOT-Platform over MQTT. 


Therfore, 

we only needed an IOT Platform with a visual appealing Dashboard, that supports our desired widgets. To find and try exactly that should be my and Jacks Task for the next two weeks.  


Starting Point:

We were given a list of IOT Platforms/Dashboards to start with:

Link to give list of IOT Platforms and Dashboards

One hour we searched through this list and tried all the different platforms. Some of them were just too expensive and did not offer somewhat like an educational version. Others seemed too complex to use and implement in the given time. Just a minority of them had support for GPS. None had it for a camera feed. 

In the end, Jack and I came to the same conclusion. We decided to look more into Wolkabout. The reasons for that:

  • looked visual appealing
  • had some form of maps integrated
  • the dashboard looked clearly laid out
  • had an android app
  • had an educational version


Wolkabout:

First steps

I started off by registering a free account and watching the introduction videos from Wolkabout on YouTube. These videos were very straight forward. It was easy to follow and visualize what to do. 

Link to Playlists of Wolkabout on YouTube


I gave the first steps a go. Creating device groups and devices worked fine. Configuration options were much available but clear. 

In the next step, I had to connect a device to the just build infrastructure. Unfortunately, I had no device to connect, so I needed to set up a dummy interface. Wolkabout did not provide such an option. I had to figure it out myself. 

Dummy device

MQTTLens and Mosquitto

This is where things got messy. I searched for ways to understand, setup and connect MQTT dummies to an IOT-Platform. I will not get into detail about this process. If you want to understand MQTT as well as make a quick and dirty setup, take a look into the links below.

Link to understand MQTT

Link to set up MQTTLens for devices

Link to set up Mosquitto as broker

After some time of research I found a solution, possible to set up in the given time. I used the Chrome Add-on MQTTLens for devices and the Windows CMD Tool Mosquitto as broker. After setup, MQTTLens could connect created devices to the broker. 

Unfortunately, the broker/devices could not connect to Wolkabout. I read through the documentation of Wolkabout and searched through YouTube and Google. I tried different settings for the dummy devices on MQTTLens, looked into the different commands of Mosquitto and what they do, and created device groups on Wolkabout with different protocols. Whatever, I could not find any further information what I could do to establish the connection, even after hours of research.

To this day, I can't say for sure why it did not work.

I gave up to use MQTTLens and Mosquitto and searched for more coding related tutorials, focused on Wolkabout itself. I found the GitHub project from the link below. 

WolkConnect-Pyhton

Link to GitHub Project WolkConnect-Python

I quickly set up a virtual machine using Ubuntu on Virtual box for security reasons. After the download of the repository and additional necessary software on my Ubuntu, I started to follow the installation instructions. This journey was soon over, after the initial build process failed. Again, I tried different settings, using the root user, etc. In the end, I could not get it to run. 

I spent the next hours with research, but the resources were scarce. I also had attempts to connect MQTTLens and Mosquitto with Wolkabout on the off chance. Slowly accepting my defeat, I decided to search for IOT-Platforms that support educational projects. 

Google IOT Core

In a few minutes, I stumbled over Google IOT Core. I wrote them our situation and brief description of infrastructure needs and finished for this day. 

Luckily, I got an answer on the next day.  

 

Current Situation:

Right now, we have an appointment with Google IOT Core on the 22. February. I hope they can support us with this task.


My Recommendations:

I hope that others in the team maybe stumbled over something that we can use. Adafruit looked very promising, for example. 

If those don't work too, I see two possible paths we could take. Maybe at the same time: 

Path one

My first idea is to stop searching for IOT-Platforms/Dashboards itself, but for similar projects like ours, based on the connections between the electronics. Perhaps we find a platform that would work, a setup around it, or something completely different. Either way, I think we need to change our point of view.  

Path two

The second idea is to find a backup-plan. The current plan is to prepare an infrastructure to create interactive dashboards with MakeCode. I would recommend implementing the interactive part on a higher level: from infrastructure to application. 

We may prepare a dashboard on our own, but stick with interactive widgets. These widgets could be: 

  • A simple manual controller GUI interface for USV ( and maybe for the drone ) 
  • MakeCode based route planning for USV (and maybe for the drone )
  • A widget for the camarafeed
  • some data diagrams for analog/digital input/output
The main advantage of this plan is that we could also create our own IOT Platform. Therefore, it would just need support for MQTT. 
The main disadvantage is: We would need to write extensions for MakeCode, GPS and camera feed, probably on our own. 

This plan is the "harder plan", but also scalable. It hardly depends on the time we have left.    

Additional Idea

In addition, we could outsource the camera feed from the whole project. The majority of drones use Mobile App clients like the one in the link below.

Link to Holy Stone drone mobile app

Some of these drones can keep a connection for multiple kilometres. This would be reasonable, because the implementation of the camarafeed on the IOT-Platform would probably be one of the greatest challenges. 

Comments

Post a Comment

Popular posts from this blog

How to: Tello drone to micro bit - Victor Alikberov, Blog#1

Image
1. Introduction. The purpose of this blog is to explain everything that is needed in order to connect tello drone to micro bit. It will cover wiring,  javascript code behind custom created tello extension and how to use this extension to connect and drive tello via microbit. 2. Wiring. Since tello drone creates its own wi-fi hotspot, additional wi-fi module is required in order to establish wireless connection between micro bit and tello. ESP8266 will be used to connect micro bit to tello wi-fi hotspot and also to establsih UDP communications between micro bit and drone. The pinout of the ESP8266 chip is shown below. ESP8266 chip pinout. Receiving port (Rx) is connected to transmitting port of the micro bit (pin 2) and transmitting port (Tx) of the ESP8266 is connected to receiving port of the micro bit (pin 1).  3.3 volts power supply should be provided both to micro bit and ESP8266 chip. Chip enable port is also connected to 3.3v power supply.  ESP8266 plus microbi...
Read more

How To: Tello Drone To RoboRealm - Jack Pender, Blog #8

Image
  Introduction: This blog post is a tutorial intended to give you everything you need in order to start streaming the camera feed from a Tello drone and get this stream into RoboRealm. All of the necessary files are included in this repository:  https://github.com/jackpender/TelloToRoboRealm I have detailed the steps but I suggest also taking a look at the repository as you go, especially the python file if you aren't familiar with OpenCV. Python File: ·         The first step is to import all the libraries that we will use. ·         We will be livestreaming the tello camera feed to the localhost using flask, so we now initialize this flask app. ·         We also open up the connection to the drone; me = Tello(), setting its name to “me”. ·         We then create a function called “gen_frames” to continuously grab the image that t...
Read more

Colour Tracking With OpenCV - Jack Pender, Blog #4

Image
  Introduction: As part of this project, the Tello drone is intended to locate anyone in danger in the water and transmit this location for the USV to travel to so it can perform a rescue. Having accessed the camera feed from the Tello and displaying it where it can be monitored, the next step was to have it implement some form of tracking. For test purposes it was decided to detect the colour green. OpenCV: The colour tracking was to be implemented in OpenCV and so I simply had to edit the python file that was running the livestream. The first step of this was to import the numpy library as this would be used to calculate the area of the detected green object. This library would also allow me to set the range of colours that the Tello was to detect. It does so by creating an array of RBG values ([22,60,200] and [60,255,255] in my case) that OpenCV can use to detect these specific colours and everything that falls in between. Here I am using these values as the brightest and ...
Read more