Close
0%
0%

Robot Communications Module

Communicate with your robot over long-ranges using a variety of plugins (RF/cellular/laser)

Similar projects worth following
How can we communicate with our robots over long distances (>1km)? Could we make a communications module that will easily allow any robot (from a little raspberry pi robot upwards) to become an avatar robot!? Even a VR avatar robot! At least we can RX video and telemetry from the robots, and TX commands to them?

This is my idea for the robotics module challenge of the hackaday prize 2018! Really it has come about because I wanted to turn a raspberry pi robot into a robot avatar that I could let out of my door and travel around 'in' through the streets!

Initial goals for the project:

1. Transmit live video from robot avatars to the base stations over long-distances (1km from base station) using cellular and RF

2. Transmit commands to the robot avatars from the base stations (e.g. to turn/stop override autonomous movement). So we can do this via RF with a 433Mhz Serial Transceiver too.

3. Allow a variety of communications methods to be connected to the communications modules as plugins (e.g. cellular, RF, laser)

4. A simple application UI to combine video received from robot with movement control command sending. This could be as a webapp too. Using a gamepad for movement control would be good.

Since it is a module for robots, I don't want it to be a main part of the robot! The robot station part of the module will be able to communicate with the robots using their wifi chips. In fact, the robot stations need not even be mounted on the robots. They just need to be in proximity, if your robot is going to be operating in a fixed area in range for wifi.

PLUGINS FOR COMMUNICATION METHODS

1. Cellular (4G/5G)

2. RF

3. Laser

PROBLEMS

Well the main problem I can think of so far is transmitting video via RF. T 

For city environments, in which we have 4G/5G, we could transmit HD video from the robots and even have VR for the UI. So giving VR robot avatars that could explore all around the city! If the robot had hands, we could even give haptic feedback via the UI! So that's where I will start, and move on to the more difficult RF stuff next, and even laser!

UPDATES:

I'll put my updates here in chronological order from the start! There are going to be many changes of mind when I learn more, and as I go through prototypes! 

Prototype for the robot station: https://hackaday.io/project/109378-robot-communications-module/log/133427-robot-station-design

RF comms plugin https://hackaday.io/project/109378-robot-communications-module/log/135300-design-of-comms-plugin-rf

31/03/2018: --RF plugin-- I've decided to use 2.4Ghz for the video transmission on the RF plugin, and use 433Mhz [serial] to tx control commands to the robots and rx telemetry. So both of these will be incorporated into the plugin. I should clarify that we expect to use RF in non-urban environments. Thus, there will be no clutter on 2.4Ghz, no reflections from buildings, and we might even have LOS (line-of-sight) to the robots! I think also, I will have to make repeater stations, if we want to get data rate of 500Kbps+ for video, when our robots are several km away from the base stations! 

06/04/2018: --RF plugin-- Now, I'm thinking of using WiFi for video transmission https://hackaday.io/project/109378-robot-communications-module/log/139889-wifi-for-video-transmission-on-rf-plugin-wlan-which-antenna

  • RF Modules: Synapse RF220x ( IEEE 802.15.4)

    Neil K. Sheridan04/07/2018 at 20:37 0 comments

    20 dBm

    2.4GHz

    RF220SU/SU-EU TX Up to 3 miles (4.8km) at 250 Kbps using a 5.5dBi antenna

    * Since it is only 20 dBm, you can increase gain on antenna to at least 9 dBi?

    Datasheet: http://www.electronics-chips.com/c/part/synapse-wireless/RF220SU-EU#Pdp-datasheet

    Datasheets and firmware for the RF220 series: https://developer.synapse-wireless.com/modules/rf220/downloads.html

    * LR-WPANs https://en.wikipedia.org/wiki/IEEE_802.15.4

  • WiFi Modules: SparkLan WUBA-171GN

    Neil K. Sheridan04/07/2018 at 20:19 0 comments

    21 dBm

    Details/datasheet: http://www.sparklan.com/p2-products-detail.php?PKey=1338-NipRzMlbr1zsg_tkbOrTX4Deewu385mp1cEyA&WUBA-171GN

    Chipset: Atheros AR9271 using https://wireless.wiki.kernel.org/en/users/Drivers/Atheros (Linux)

    Pinouts:

    1 WPS function ; Active “High” 

    2 VCC (+5V)

    3 USB_D- 

    4 USB_D+
    5 GND


  • WiFi Modules: Skylab SKW77

    Neil K. Sheridan04/07/2018 at 19:29 0 comments
  • WiFi for video-transmission on RF Plugin (WLAN) / which antenna?

    Neil K. Sheridan04/06/2018 at 19:09 0 comments

    Can we use WiFi to transmit video long-distances? What power will the receiver get for various distances/antenna gains?

    How about we use WiFi to transmit video and telemetry via WLAN from the robot station to the base station? Can we use WiFi over long-distances?

    First I thought about using a high-power WiFi module on the RF plugin. So I found this one to start, although there are plenty of others:  SkylabHigh power module SKW77. This has transmission power of up to +27dBm (so ~500mW)[1] and frequency range is 2.400GHz—2.4835GHz. IEEE 802.11b/g/n 2×2 300Mbps. IPEX connector (Hirose U.FL?) for antenna.

    The 5 dBi antenna is the Antenna +4-5dBI 2.4GHz/5.1GHz/5.8GHz 90Deg SMA Rev 

    * Remember that dBm is logarithmic scale!

    Then what kind of antenna should be used? Here we have two 5dBi antenna. We can calculate the power received by the base station by entering the following into the Friis Equation[2]:

    • Transmit Power (Pt) : 27 dBm
    • Transmit Antenna Gain in dBi (Gt): 5 dBi
    • Receiver Antenna Gain in dBi (Gr): 5 dBi
    • Wavelength: 0.12491352 (for 2.4GHz)

    If we have:

    Antenna Separation in metres (R) = 1000, that gives -63 dBm at receiver 

    If we have:

    Antenna Separation in metres (R)= 500, that gives -57 dBm at receiver

    Ok, so -63 dBm and -57 dBm are both great for transmitting video!

    What happens if we increase the antenna separations (R)?

    Antenna Separation in metres (R) = 1500, gives -66.5 dBm at receiver

    That's still ok for transmitting video! 

    Antenna Separation in metres (R) = 2000, gives -69 dBm at receiver 

    That's not so good for video now. Since we wanted better that -67 dBm

    What next?

    This is just great! But it is all based on ideal conditions! No interference, no reflections, etc. It would certainly be great for robots in a desert, or going along the beach (e.g. trash picking robots!). I.e. in situations when we have LOS. But it wouldn't be so great for robots in a forest! And it would be no good at all for robots in a city! But of course, we would use 4G/5G for those robots.

    Well, we can't just increase the transmission power, because there are regulations on this! And we do have to worry about our batteries too! These are the EIRP limits (https://en.wikipedia.org/wiki/Effective_radiated_power) . I only looked at this briefly, but it seems FCC (USA) rule for FCC 2.4 GHz BAND (POINT-TO-MULTIPOINT) is 36 dBm EIRP limit. 

    One thing we can do is to increase the antenna gain from 5 dBi to 9 dBi. Keeping the power at 500mW. That should be inside the regulations! And would give us power of around -67 dBm with antenna separation of 4000 metres! Example antenna being ANT-2G-OM-9-N-V (~£22)

    Another thing we can do is increase the gain on an receiver antenna! We could go really high with the gain on this! For instance, TP-Link TL-ANT2424B 2.4GHz 24dBi Grid Parabolic Antenna (~£70) gives 24 dBi gain!! But wait, won't this break the EIRP limits? It would do, if it were to transmit! But does it have to transmit? Well it has to transmit as 802.11 specifies bidirectional, and we are going to have problems with security if we don't transmit too! BUT! There have been some exploits of the ESP32 to make it just a one-way data stream to the receiver! I don't want to go into them, because they could be used for bad purposes :-( . Maybe we could RX on the 24 dBi antenna (for video), and TX on a 9 dBi antenna to conform to 802.11 standard? Ok, that doesn't sound like the best idea! But it sure would give a long-range! And is worth investigating!

    So with an antenna gain of 24 dBi on the receiver, and 9 dBi antenna gain  on transmitter, we could get power of  -66 dBm at the receiver with antenna separation of 20km! 

    [1] Converting dBm to Watts

    P(W) = 1W ⋅ 10(P(dBm) / 10) / 1000 = 10((P(dBm)- 30) / 10)

    [2] Friis Equation

    https://en.wikipedia.org/wiki/Friis_transmission_equation

  • 5.8Ghz: Video transmitters used for FPV in drones, and possible modules

    Neil K. Sheridan04/05/2018 at 19:57 0 comments

    Following on from the 5.8Ghz log, I'm investigating some of the video transmitters used by drones.

    MatekSys  VIDEO TRANSMITTER 5.8G VTX-HV W/ BFCMS CONTROL

    • Input voltage: 7~27V(2~6S LiPo)
    • 40 standard channels, A/B/E/F/R Band
    • Output Power: 25, 200, 500mW switchable
    • BEC Output: 5V/1A for Camera or FC

    You can see details here http://www.mateksys.com/?portfolio=vtx-hv#tab-id-4

    TBS UNIFY 5G8 PRO HV

    RACEDAYQUADS MACH 2


    These are all fairly low-cost (the VTX-HV is around £30), but they are not open-source, so I don't know the components used etc. But I'm sure we can have a look at them anyway! Even if they are not going to be used in the project!


    POSSIBLE 5.8Ghz MODULES FOR USAGE IN PROJECT

    Anyhow, just looking around, there are some 5.8Ghz TX modules that can be incorporated into our RF plugin for the robot station. We'll be making our own PCB with this and the 433Mhz module on it! We can't just use one of these video transmitters from drones! Because they are not open-source!

    FX756T 5.8G 600mW 32CH 

    So yes, a lot of the 5.8Ghz modules are lower power than this (600mW), so wouldn't be much good for our goal of 1km! This one seems promising!

  • 5.8GHz for video transmission

    Neil K. Sheridan04/05/2018 at 18:50 0 comments

    So, I went to investigate how drones transmit video! They use FPV modules/systems (first-person video). Many use 5.8Ghz. So they might use 200 milliwatt, 5.8GHz module to transmit video 2.4km range. Or 800 milliwatt for 4.8km range. Note that drones broadcast an omnidirectional radiation pattern, using circularly polarized omnidirectional antennas on transmitter and receiver. We don't need that for the robots, since we are in just one direction from them. Maybe we could use cast surface grid antenna, or dish antenna, on the base stations and robot stations?

    I've found a video intro to FPV, and antenna type, dbi/gain, power, diversity controllers, here! 

    Remember, like I said, the robots can use a narrow beam vs. omnidirectional that drones need - so we can get much better range without same power costs. The idea of a diversity controller is quite interesting! We have this attached to the base station. It would have different antenna on it, and it will select the antenna getting the best signal. This is the one it would switch to the base station. So it could have a circular polarised antenna on it, and a dish antenna, and a patch antenna, and it will select which is getting the best signal! Drone users are making even 6-channel diversity controllers! Something I'll look into! The next step up for drones, is to use motorized tracking antenna that will point at the drone. These can give up to 115km range for video transmission e.g. http://www.uavfactory.com/product/86 !!

    Ok, so I will look at some of the ~5.8Ghz FPV/video transmission modules that have been designed for drones next! In fact, I have come across an open-source diversity controller. You can see it here on github https://github.com/sheaivey/rx5808-pro-diversity / so it uses the RX5808 5.8G 8CH receiver module

  • TCP vs UDP protocol over RF

    Neil K. Sheridan04/03/2018 at 20:06 0 comments

    Can we use TCP over RF? Is there going to be an issue here about overloading the sender with NAK or NACK (negative-acknowledgement [ACK]) signals because we are getting many lost packets/packets out of order/etc.?

    I found a white paper on this issue here: http://www.rtlogic.com/~/media/datasheets/whitepapers/rtl-mwp-079_adapting%20rfif%20over%20ip%20within%20range%20applications.pdf

  • Design of Comms Plugin: Laser

    Neil K. Sheridan03/30/2018 at 19:33 0 comments

    Well, we are going to need line-of-sight (LOS) for this! And hope that birds won't fly through the laser!

    Anyway, I'm just gathering some useful research on this at the moment:

    10m/500Mbps WDM visible light communication systems https://www.osapublishing.org/oe/abstract.cfm?uri=oe-20-9-9919

  • Design of Comms Plugin: RF

    Neil K. Sheridan03/30/2018 at 18:32 0 comments

    Introduction

    [31/03/18]

    I've decided the RF plugin should use 433Mhz for serial communication: in order to transmit control commands from base station to robot, and transmit telemetry (GPS, temperature, battery voltage, etc.) from robot to the base station. This is fairly simple, so I won't list the modules.

    The RF plugin should use ~2.4Ghz for video transmission from robot to base station. Unfortunately, there is a lot of stuff using 2.4Ghz so there will be interference depending on area. 

    I was thinking to try the Synapse RF266PC1 first. This comes with Synapse SNAP network operating system [pdf], an embedded Python virtual machine . So you don't need an MCU on the module. Here's the SNAP SNAPcore firmware primer [pdf]. BUT this has been discontinued! So, instead, I will try the Synapse RF220SU-EU instead. This seems even better! With 4.8km LOS range giving 250Kbps. 

    I doubt we are going to get much of range in urban environment at 500Kbps non-LOS! But let's not be put-off, because we would likely have 4G there instead! The RF comms is going to be used in environments that don't have 3G/4G/5G! So probably not many buildings in these places, which means we might even have LOS. And nothing else using 2.4Ghz either!

    2.4Ghz Modules

    Here are some of the 2.4Ghz modules I've found as candidates for this!

    Synapse RF220SU-EU

    4.8km range [this is actually LOS outdoors at 250Kbps]

    2.4Ghz band operation

    250Kbps, 500Kbps, 1Mbps, 2Mbps data rates

    Transmit power: +20 dBm

    150mA TX at +20 dBm 

    22mA RX

    U.FL Connector / rp-sma connector for antenna

    Datasheet: http://www.synapse-wireless.com/upl/downloads/resources/supporting-documents/product-brief-rf220uf1-and-rf220su-b874d350.pdf

    Nordic Semiconductors nRF24L01+ Single Chip 2.4GHz Transceiver

    100m range 

    2.4GHz ISM band operation

     250kbps, 1Mbps and 2Mbps on air data
    rates

    11.3mA TX at 0dBm output power

    13.5mA RX at 2Mbps air data rate

    Datasheet: https://www.sparkfun.com/datasheets/Wireless/Nordic/nRF24L01P_Product_Specification_1_0.pdf

    [there are supposedly lots of fakes for this chip!]

    Synapse RF266PC1 - 2.4GHz (chip antenna)

    1200m range (at 250kbps)

    so I assume we can get to 2Mbps at shorter ranges

    2.4GHz ISM band operation

    250kbps, 1Mbps and 2Mbps on air data
    rates

    Transmit Current (Typ@3.3V) 130mA [TX]
    Idle/Receive Current (Typ@3.3V) 25mA [RX]

    Datasheet: https://cdn.sparkfun.com/datasheets/Wireless/General/Synapse-RF266PC1-Engine-Data-Sheet.pdf

    [possibly discontinued - see the latest modules here http://www.synapse-wireless.com/upl/downloads/library/rf-modules-comparison-chart.pdf ]

    TANGO-24PA-PCBANT -  RADIO TRANSCEIVER, SPI, 2MBPS

    1km range 

    2.4Ghz ISM band operation

    250Kbps / 1Mbps / 2Mbps  data rates

    Transmit  100mW (+20dBm) Maximum Transmit Power

    Datasheet: http://www.farnell.com/datasheets/2267055.pdf

    MRF24WB0MB/RM -  RF Transceiver Radio Module, 802.11B, U.Fl Connection, 2.483 GHz

    [IEEE 802.11 compliant WiFi transceiver module]

    400m range

    2.483GHz operation

    1Mbps/2Mbps data rates

    Designed for use with PIC18, PIC24, dsPIC33, and PIC32 MCU's with downloadable TCP/IP stack

    Datasheet: http://www.farnell.com/datasheets/1738902.pdf

     As you can see, we won't be able to transmit HD video over long-distances (1km+) using 2.4GHz. There is nothing to stop us implementing repeater stations however! Exploration robots could deposit their own repeaters as they progress I suppose! All of these modules are low-cost (<£30). 

  • Design of Comms Plugin: Cellular

    Neil K. Sheridan03/28/2018 at 21:12 0 comments

    Design of comms plugin cellular / software library

View all 12 project logs

Enjoy this project?

Share

Discussions

Ulysse wrote 04/01/2018 at 19:16 point

A project that I follow with attention of course

  Are you sure? yes | no

Similar Projects

Does this project spark your interest?

Become a member to follow this project and never miss any updates