INAV 1.8 is just around the corner

Next release of INAV, navigation enabled flight controller software, is almost here. INAV 1.8 RC1 has been released just 2 days ago. List of changes is rather long, so here is shortened version with the most important things:

  • STM32F1 boards like Naze and CC3D are no longer supported. INAV 1.7.3 is the last version that can used on those boards. All F1 users are encouraged to migrate to F3 or F4 boards
  • INAV is now able to get current time from GPS. Time is saved in blackbox logs and can be displayed on OSD
  • SmartAudio (TBS Unify Pro video transmitters) and TrampHV video transmitter support. When connected over free UART (TX pin only) and properly configured, band, channel and output power can be changed using OSD and CMS subsystem. Changing this over MSP and SmartPort on a OpenTX radio (Taranis, Taranis Q X7) with Betaflight LUA script is not tested yet and probably does not work (yet)
  • Receiver type is no longer selected using feature command. There is receiver_type CLI variable instead. Configurator handles this in transparent way
  • Multiple OSD changed. And by multiple, I really mean multiple. Not only is has better update rate but also takes less memory and less CPU time. On top of that, new OSD elements has been added as well:
    • Combined "On time"/"Fly time"
    • System messaged indicator showing additional modes information, arm failure reasons, navigation stages
    • Average cell voltage
    • New throttle indicator showing also throttle requested by navigation subsystem, not only by user
    • Time indicator
    • Heading graph indicator
    • VTX band and channel, required SmartAudio or TrampHV
    • Distance alarm
    • Negative altitude alarm
  • AUX channels have been remapped to RC channels. AUX 1 is now CH 5, AUX 2 is CH 6 and so on. Be careful when restoring rc map from dump!
  • Navigation modes override MOTOR_STOP. This solves the problem of motors shutting down in rare cases when Navigation modes were enebled
  • Several other changes and bugfixes

Read More

And the most popular flight controller for INAV is… (August 2017 edition)

More than 2 months after previous “What it the most popular flight controller for INAV“, welcome to second edition.

I mportant, this is not the number of boards flashed with INAV, but rather number of times a board was connected to Configurator!

This counts TARGET software name, not retail name. For example, all clones of Naze32 will be counted as Naze32

Data was taken in August 2017, multiple connections during single user session are stored as single entry.

  1. SP Racing F3 – 25% of all boards, lost 4 percent points
  2. Omnibus F4 Pro (aka v2 aka SD) – 15% of all boards, gained 5 percent points
  3. Naze32 rev. 5 and 6 – 13% of all boards, lost 6 percent points
  4. Omnibus F3 – 12%, lost 1 percent point
  5. Naze32 older than rev. 5 – 7%, gained 2 percent points
  6. Omnibus F4 v1 – 4%, gained 1 pp.
  7. Omnibus F4 v3 and v4 – 3%, first time in TOP 10
  8. SP Racing F3 EVO – 3%, lost 2 pp.
  9. CC3D – 3%, lost 1 pp.
  10. AnyFC F7 – 1%, first time in TOP 10

Other targets have 14% in total.

This data also says something about CPU generations distribution:

  1. STM32 F1 – 27%
  2. STM32 F3 – 47%
  3. STM32 F4 – 25%
  4. STM32 F7 – 1%

What changed comparing to June?

  1. First F7 board appeared in TOP 10!
  2. F1 and F3 targets are loosing market share, while F4 and F7 are more and more popular
  3. Naze32 is no longer second most popular board
  4. OpenPilot Revolution and Airbot F4 (more less the same) are out of the TOP 10
  5. Boards with integrated OSD are more and more popular. They already have 47% of market share

Read More

How to connect APM Airspeed Sensor (Pitot Tube) to INAV

Recently released INAV 1.7.2 introduced new hardware support: analog airspeed sensors. I’ve written new but it is new only to INAV. Analog airpseed sensors aka APM Pitot Tubes aka Ardupilot Airspeed Meter and one the market for years. And they are cheap. By cheap, I mean below $30. And until recently they were usable only in APM world.

Pitot Tube for INAV Continue reading “How to connect APM Airspeed Sensor (Pitot Tube) to INAV” »

Read More

VL53L0X Lidar is… disappointing

I just finished first test flights with laser rangefinder VL53L0X and well, I'm disappointed. Before I will show any data, let's review test conditions:

  • VL53L0X configured for Long range mode: 2m
  • 8PM, obscured sky, no direct sunlight
  • Over concrete

And it looks like this:

vl53l0x altitude graph

vl53l0x altitude graph

Solid readout ended at more less 80cm. Above that, valid readout appeared only from time to time. So I have to assume, that everything above 1m is absolutely not reliable and useful range is max. 75cm. Probably much less over different surfaces.

And that means it will have rather limited usefulness. Landing detection probably. Or low altitude flights with terrain following. I tried that too and it worked as long as quadcopter was not higher as 20 or 30cm above ground.

Oh well… I had high hopes for VL53L0X. Next attempts will be with hybrid approach: US-100 sonar and Sharp optical (IR) GP2Y0A710K0F (up to 5 meters) data fusion.

Read More

Bye bye Naze, we will not miss you that much

I have a news for all INAV pilots using Naze32, Flip32 and other boards compatible with NAZE target. You might call it a bad news, but reality is that it is not that bad and was long anticipated. It is official: INAV 1.7.2 was the last INAV release with NAZE target. That means the following: INAV 1.8 will not be available for Naze32, Flip32 and other boards compatible with that target.

You might was "why?". Quite simple: not enought flash memory, no way for new features to fit in. There was even not enought flash memory for bugfixes. And to be honest, I do not remember last time when NAZE users really got a new feature. Almost all new things were disabled for them. For more than a year, compiling NAZE target after adding something new was quite a challenge.

Does that mean that you can not use Nazes any more? Absolutely not. They are good boards and INAV 1.7.2 works on them just fine. You only will not be able to upgrade to INAV 1.8. And trust me, it is really worth it to invest in something better like F3 board. They are not that expensive after all…

By the way, CC3D is the next thing to be removed. Not yet, it still fits flash. Barely, but fits…

Read More

Large wooden BiCopter

You just know a great project when you see one. Like, for example, large BiCopter made from wood and running INAV and flight controller software that kermets has built.
BiCopters might look strange, but the idea is quite simple. Instead of 4 motors, we have 2 motors and 2 servos tilting the motors. Like a one step further from a tricopter.

Similar idea (not the same, only similar) is used on large helicopters like Boeing CH-47 Chinook

Read More

INAV jumped over the bench

Let's say I have almost a good news about sonar support in INAV: yesterday I flew terrain following mode with experimental INAV code. And did not crashed when shooting video below. I did crashed next code version, but that is only a minor detail, right?

  1. When shooting that video I did not touched throttle stick. Altitude control was 100% automatic
  2. It is US-100, not HC-SR04 ultrasonic rangefinder!
  3. US-100 was connected to Omnibus F4 Pro using experimental I2C interface with ATtiny85
  4. If you want to try it, here is the code. But be prepared to crash 🙂

Read More

If not HC-SR04 then what? US-100 maybe….

A week ago I wrote that HC-SR04 ultrasonic rangefinder is crap and it is still true. Only around 1m useful range and very jumpy behavior above are at least problematic…

There is a better solution. US-100 ultrasonic rangefinder. Comparing to HC-SR04, it has at least twice the useful range. My preliminary tests gave solid readout over concrete up to 2 meters and something. Let's call it 2 meters. But what more important, US-100 was reporting out-of-range state instead of some random values when being out of range! Huge step forward comparing to HC-SR04.

us-100 is much better than HC-sr04

Unfortunately, better does not mean good. US-100 also has some problems.

us-100 is much better than HC-sr04  ut still not good

Readouts becomes "jumpy" at the edge of useful range and they are far far away from declared. But OK, declared always means "perfect conditions". Still, trust me on this, comparing to this, HC-SR04 is crap…

Read More

INAV 1.7.2 has been released

“Let’s do the news….” and INAV 1.7.2 has been released yesterday. Besides new targets (MATEKF405, Alienflight F7, SP Racing F3 NEO) there are few quite important changes:

  • ADC channel to function mapping is now configurable. Any ADC using function (battery voltage, current, RSSI) can be assigned to any ADC pin. You smoked Current pin? No problem, now you can use RSSI for that. It can be done using vbat_adc_channel, rssi_adc_channel, current_adc_channel, airspeed_adc_channel CLI commands
  • Support for analog pitot tubes, based on MPXV7002DP, known as APM Airspeed Sensor. It requires some hacking, but is pretty simple. I will publish detailed instructions in a next few days
  • Servo handling improvements:
    • servo min and max now do what they are supposed to be doing: output is scaled to reach them on max/min input, not just clip. And that means, that aileron differential is not super simple to archive
    • servo rule speed is now defined as 1 speed = 10 us/s. So, speed 100 means that full servo sweep will be done in 1 second. speed 50 means full servo sweep in 2 seconds and so on
    • smix min and max parameters were removed
  • Support for eLeReS receivers built into KFC flight controllers
  • Total flight statistics via stats, stats_total_time and stats_total_dist CLI commands
  • AnyFC F7 improvements:
    • buzzer on output #9
    • SD card detection is now configurable via sdcard_detect_inverted
    • support for external barometers using ANYFCF7_EXTERNAL_BARO target
  • Fixed wing landing after RTH that I already described here

Full list of changes is available here.

Read More