tmcadam / mlrs

2.4 GHz & 915/868 MHz & 433 MHz/70 cm LoRa based telemetry and radio link for remote controlled vehicles

mLRS

The mLRS project offers an open source 2.4 GHz & 915/868 MHz & 433 MHz/70 cm LoRa-based high-performance long-range radio link, which provides transparent bidirectional serial connection combined with full remote control.

The 'm' in the project name alludes to 'Mavlink', as it will have features which optimizes performance for Mavlink systems. However, it always will also provide a transparent serial link and hence will be of wider use and by no means limited to Mavlink systems only. The 'LR' in the project name alludes to 'long range', which however should not be understood in terms of an absolute range, like 100 km, but - of course - as the best possible range under the given conditions. Physical laws simply say that the higher the data rate the shorter the range. So, mLRS cannot compete range-wise with systems which achieve their range by reducing data rate to the minimal, at the cost of the associated compromises. The goal of mLRS is to achieve a high range under the condition of a relatively high data rate.

Typical specs could be 'plenty' of full-resolution RC channels, with 50 Hz update rate and serial data rates of about 3-5 kBytes/s at 2.4 GHz.

Many LRS or radio links with telemetry exist, among them open source projects such as SiK radios, OpenLRS, ExpressLRS, but also befinitiv wifibroadcast based projects like OpenHD or Ruby, closed source hobbyist projects such as UltimateLRS, QczekLRS, as well as commercial systems such as DragonLink, RFD900, Dronee Zoon, Siyi, and also TBS Crossfire and alike.

However, while all these systems are truely excellent and achieve their goals, and some of them are indeed close to what the project aims at, none of them checks all boxes, like

• relatively cheap
• 2.4 GHz, 915/868 MHz, 433 MHz/70 cm
• LoRa
• full-duplex serial link with sufficient data rate
• plenty full-size RC channels
• open source
• rich features and outstanding performance for Mavlink systems

The mLRS project fills this gap.

Disclaimer

You of course use the project fully at your own risk.

Project Status

The project is work in progress, as there is still plenty of room for ideas and improvement.

The essential features, however, i.e., the RC link and the serial (MAVLink) data link, are stable and robust, and perform quite well.

The mLRS system also provides a high level of usability such as a variety of options for input/output, parameter setting via the mLRS transmitter, optimization for ArduPilot/PX4 systems, wireless connection to ground control stations like MissionPlanner or QGC, or support of the Yaapu telemetry app without extra hazzles. It also integrates well with the mTX (formerly MAVLink for OpenTx) project, which yields a most fluid user experience.

It supports the SX1280/1, SX1276, SX1262 and LLCC68 Semtech chips, and thus the 2.4 GHz, 915/868 MHz and 433 MHz/70 cm frequency bands.

It provides 16 RC channels with the following layout (layout is equal in all operation modes):

• CH1 - CH8: 8 channels with 11 bit resolution (CH1 - CH4 have a higher reliability margin)
• CH9 - CH12: 4 channels with 8 bit resolution
• CH13 - CH16: 4 channels with three steps (CH13, CH14 have a higher reliability margin)

It provides these operation modes:

	50 Hz	31 Hz	19 Hz	FLRC (111 Hz)	FSK (50 Hz)
frequency bands	2.4 GHz	2.4 GHz 915/868 MHz 433 MHz/70 cm	2.4 GHz 915/868 MHz 433 MHz/70 cm	2.4 GHz	915/868 MHz 433 MHz/70 cm
chip sets	SX128x	SX128x, SX126x/LLCC68	SX128x, SX126x/LLCC68, SX1276	SX128x	SX126x/LLCC68
downlink serial rate	4100 Bytes/sec	2562 Bytes/sec	1547 Bytes/sec	9111 Bytes/sec	4100 Bytes/sec
uplink serial rate	3200 Bytes/sec	2000 Bytes/sec	1207 Bytes/sec	7111 Bytes/sec	3200 Bytes/sec
receiver sensitivity	-105 dBm	-108 dBm	-112 dBm	not for LR	not for LR

Further features:

• full diversity: mLRS transmitters and receivers which feature two Semtech Lora chips provide full diversity, for both receiving and transmitting. This really improves link quality in the far range, and allows advanced dual-antenna setups on the transmitter side.
• the receiver parameters can be set from the mLRS transmitter or radio; no need to mess with the receiver in any way.
• the transmitter and receiver parameters can be set via a Lua script, a CLI, or an OLED display. There is no need to recompile or reflash the firmware in order to change an option or parameter setting.
• bind mode for binding "unknown" receivers to a transmitter.
• the mLRS system can be configured through the parameters for a wide range of applications and use cases. For a pictoral representation of some typical examples see mLRS Setup examples, and for more details Documentation.
• 10 model configurations stored in the mLRS transmitter, selected by "Receiver" number in OpenTx/EdgeTx radios.
• support of CRSF and ArduPilot passthrough protocol; enables using the Yaapu Telemetry app on standard radios (out of the box, no need for extra dongles!).
• optimizations for ArduPilot and PX4 autopilot systems.
• technologies introduced by mLRS:
  • innovative flow control for smoother and more robust data flow
  • MavlinkX for reduced packet loss and data compression
• "except" and "ortho" features
• support for buzzer, OLED display & five-way button, serial2.
• support of ESP32 modules for wireless connection to a ground control station.
• support of plenty platforms: STM32F103, STM32G4, STM32L4, STM32F3, STM32WLE5, Wio-E5, ESP8285, E28, E22, E77, SX1280, SX1262, SX1276, LLCC68.

Community

Discussion thread at rcgroups: https://www.rcgroups.com/forums/showthread.php?4037943-mLRS-Lora-based-Mavlink-oriented-open-source-radio-link

Discord server by LELE2022: https://discord.gg/vwjzCD6ws5

Range

The range which one may expect can be estimated from the standard math; the ImmersionRc RF Link Range calculator comes in very handy here. Let's assume: power = 20 dBm (100 mW), antenna gain = 2 dBi, link margin = 12 dB (note: 12 dB link margin is conservative). Then, for the three LoRa modes:

	50 Hz	31 Hz	19 Hz
2.4 GHz	7 km	10 km	15 km
868/915 MHz	-	26 km	42 km
433 MHz/70 cm	-	55 km	87 km

For the 2.4 GHz band, the available range test reports consistently exceed the above estimated ranges (e.g., 8.3 km were reported for 2.4 GHz, 50 Hz, 9 dBm (8 mW), which translates to 29 km at 100 mW). For the other frequency bands less information is available. Note that mLRS supports full diversity, which when enabled has been found to significantly improve performance at lower link budget, i.e., allows you to operate at larger ranges.

The FLRC and FSK modes are not intended for long range.

Hardware

Hardware is still a problem. One might be tempted to think that the commercial ExpressLRS hardware should be good platforms, but this is unfortuantely not so. The ESPs they use do not offer the peripherals which are desired for mLRS transmitters, and STM32's were hence chosen as main platform. However, mLRS also supports the ESP chipset, and thus a number of ExpressLRS hardware.

The code currently supports:

• Flysky FRM303 transmitter module (2.4 GHz)
• Frsky R9M and R9M Lite Pro transmitter modules and R9 MX, R9 MM and R9 Mini receivers (868/915 MHz)
• SeeedStudio Wio-E5 Mini and Grove Wio-E5 boards (868/915 MHz, 433 MHz/70 cm)
• EByte E77 MBL board (868/915 MHz, 433 MHz/70 cm)
• several DIY boards you can find in https://github.com/olliw42/mLRS-hardware
• ExpressLRS receivers which use the ESP8285 chipset (hardware which are based on ESP32 chipset are not yet supported)

In the 915/868 MHz range, the Frsky R9M & R9 MX system provides a simple and readily available entry into mLRS. In this sense it is the best option available currently. The disadvantage of the R9 receivers is however their low transmission power (50 mW); the telemetry range is thus relatively short (still several 10 km). This can be mitigated by using the R9M Lite Pro (or R9M) transmitter module as receiver, which is supported by mLRS. A better option can be the various ExpressLRS receivers, some of which provide up to 500 mW transmission power, and some of them are cheaply available. The combination of a Frsky R9M transmitter module and a ExpressLRS receiver can be a fantastic choice. The downside of all these gear is that they only support the 19 Hz mode.

The SeeedStudio Wio-E5 boards and the EByte E77-MBL board are also readily available, and hence excellent options too to enter mLRS. The "easy-to-solder" module, which uses an Ebyte E77 module, is a simple DIY option for building a mLRS receiver (it can also be used to build a mLRS transmitter). These boards are all based on the STM32WL5E chip and thus provide all the advantages of the SX1262, like the 31 Hz mode. Their maximum power is 22 dBm, and they can be used in the 915/868 MHz and 433 MHz/70 cm frequency ranges.

In the 2.4 GHz range, the Flysky FRM303 transmitter module is a readily available, albeit expensive, option. mLRS supports using it as transmitter as well as receiver. Concerning receivers, if the full potential of mLRS is desired, the DIY options are probably the way to go (they are all based on STM32). The DIY options also offer transmitters, including the most capable mLRS transmitters available. The, by far, easiest receiver option is however ExpressLRS receivers, as they are readily available.

Don't hesitate to join the discussion thread at rcgroups or the discord channel for more details.

Firmware: Flashing

Ready-to-flash firmware can be found in the "firmware" folder. All you need to do is to flash the binary file appropriate for your board into the device (it is not required to install the software for compiling as described in the next chapter or the docs). The mLRS transmitter can then be configured to your needs via the CLI, the mLRS Configuration Lua script, or the OLED display if available. The mLRS receiver is configured by first binding it to the transmitter, and then configuring it through the transmitter, exactly like the transmitter is configured.

Software: Installation Bits and Bops

mLRS uses STM32CubeIDE for STM32 targets, and Platformio with VSCode for ESP targets. For STM32 targets this procedure should work (for ESP targets see ESP Development):

Let's assume that the project should be located in the folder C:/Me/Documents/Github/mlrs.

1. Clone and setup the project files

• open a command line processor
• cd into C:/Me/Documents/Github (not C:/Me/Documents/Github/mlrs !)
• git clone https://github.com/olliw42/mLRS.git mlrs
• cd mlrs
• run run_setup.py. This does three steps: initializes submodules (git submodule --init --recursive), copies ST HAL and LL drivers to the targets, and generates mavlink library files.
  • Note: Ensure that all three steps are executed completely.

For cloning you of course can use any other tool you like.

2. STM32CubeIDE (for STM32 targets)

• download and install STM32CubeIDE.
  • Note: Install into the default folder if possible.
• start STM32CubeIDE
• in Launcher select Workspace by hitting [Browse...] button, and browse to C:/Me/Documents/Github/mlrs/mLRS. Hit [Launch] button.
  • Note: it is not C:/Me/Documents/Github/mlrs but C:/Me/Documents/Github/mlrs/mLRS! If you proceed with the wrong path then there will be a compile error "undefined reference to main_main()"!
• in the IDE's top bar go to File->Open Projects from File System
• in the Importer select Import source by hitting [Directory...] button, and browse to the desired project. E.g. select C:/Me/Documents/Github/mlrs/mLRS/rx-diy-board01-f103cb. Hit [Finish] button.
• change from Debug to Release configuration: Go to the 'hammer' icon in the top icon bar, click on the down arrow right to it, and select Release.
  • Note: if you don't do that then there will be a compile error "undefined reference to main_main()"!
• open the file mlrs-rx.cpp or mlrs-tx.cpp into the editor
• compiling should work now: Go to the green 'right-pointing triangle' icon in the top icon bar and click it
• Repeat the last five steps for each board you are interested in

The STM32CubeIDE has its weirdness, so you may have to get used to it.

In case of issues with this procedure, don't hesitate to join the discussion thread at rcgroups or the discord channel, or submit an issue in the github repository.

Dependencies

You need to have git and Python3 installed. Depending on the Python3 distribution you may need to install further libraries.

Further Documentation

You find many more information here:

mLRS Documentation