Alt Hold, Loiter - Retro thinking for a new solution?

Don't know if this has been thought of and dismissed or not. How about using the IMU as an INS.

 

I think most of us are expecting higher accuracy than is possible while keeping cost down. We also have a problem with perception of error due to scale. Baro and GPS errors are acceptable in full scale but seem huge for our small toys.

 

My proposal is to use baro and gps only for waypoints and then switch over to INS to maintain position. This would also allow the copter to fight the wind better. The stick would then become purely rate based for speed unless in acro mode, just like FBW in APM. The IMU already has the accuracy as is evidenced by the multitude of indoor videos show a perfectly stationary copter. Sonar is still the best choice for low level ground avoidance.

 

A possible addition some time in the distant future (AC4 or5) would be to add INS based missions to allow circus tricks like flying through hoops. The board could possibly even be placed in learn mode and a short indoor mission walked through by hand so it can learn to put on a show.

Is any of this possible? Don't know. Just my thoughts.

You need to be a member of diydrones to add comments!

Join diydrones

Email me when people reply –

Replies

  • INS is nearly impossible to make really functioning with existing inexpensive MEMS IMUs. We experimented a lot. For keeping a few cm level accuracy, the time between calibration with an external system was just 1/4..1/2sec. Thus, after 1/2 sec or so, you really need to provide a location measured precisely with an external system to compensate the drift of the onboard IMU after the double integration of the drifting accelerometer data.

    We did it, thought, after 1.5 years of work. It is called IMU+Ultrasonic sensor fusion and is available in our Precise (±2cm) Indoor Positioning system in NIA configuration. It requires that ultrasonic-based location update is not less than 4Hz. When available, you can precise IMU (INS) based tracking with 100Hz update rate and latency of 12-15ms on board of a drone or robot.

    https://marvelmind.com/
    Marvelmind Robotics
    Precise (±2cm) Indoor positioning for autonomous robots, drones and humans
  • Yes I did mean just using dead reckoning. So I correctly assumed that current affordable hardware is still not up to the task. By the sounds of it maybe AC4 or 5 is still optimistic, maybe version 10 in a few years, Having been involved with the avionics research industry, I realise how long things can take. It took something like 35 years from a FBW plane to the first completely FBW helicopter with no mechanical backup. What we have now was only a dream not that long ago.

    Thanks.

  • Developer

    Any videos showing great stability indoors owe that stability to ultrasonic and infrared sensors and lack of wind.  And those videos of quads flying through hoops etc were done using external multicamera optical imaging systems costing six figures plus.  If you think the precision using baro alt is bad, wait till you try doing a double integration with our low cost accelerometers.  An accurate INS solution with this level of gyro and accelerometer sensors is not possible.  If it were as simple as designing a good Kalman filter it would have been done long ago.

     

     

  • 3D Robotics
    Can you explain a bit more what you mean by INS? I know that means Inertial Navigation System but I'm not quite sure what that includes or not. Do you mean just using dead reckoning?
This reply was deleted.

Activity

DIY Robocars via Twitter
10 hours ago
DIY Robocars via Twitter
RT @NVIDIADRIVE: Season Beta has arrived! Check out the first race of the @roborace season, livestreaming from Anglesey Circuit Sept. 24-26…
18 hours ago
DIY Robocars via Twitter
RT @IndyAChallenge: Here is the ROS-based small-scale vehicle platform showing the ROS Visualization (Rviz) screen with multiple ROS nodes…
yesterday
DIY Robocars via Twitter
RT @IndyAChallenge: Here is Black & Gold Autonomous Racing's simulated racecar running a lap @IMS and showing the team's code overtaking ot…
yesterday
DIY Robocars via Twitter
yesterday
DIY Robocars via Twitter
RT @donkey_car: New Maintainers! Donkey Car 4.0 and more! Here is our September Newsletter! https://donkeycar.substack.com/p/september-donkey-newsletter?r=2ji2t&utm_campaign=post&utm_medium=web&utm_source=copy
yesterday
DIY Robocars via Twitter
RT @chr1sa: Kinda cool that we just hit 2,500 members of our Bay Area self-driving car racing league. More than 10k globally. Fast, cheap…
yesterday
DIY Robocars via Twitter
RT @chr1sa: Our next @DIYRobocars virtual AI car race is next Saturday. Compete from home using the @donkey_car simulator -- no physical ca…
Sunday
DIY Robocars via Twitter
RT @RoboticMasters: Students from @Sydney_Uni working hard on improvements and changes to @donkey_car simulator. @diyrobocars @adafruit…
Sunday
DIY Robocars via Twitter
Practice virtual race this Saturday; the real thing will be on Oct 3 https://www.meetup.com/DIYRobocars/
Sep 23
DIY Robocars via Twitter
Sep 23
Derrick Davies liked lisa TDrones's profile
Sep 23
DIY Robocars via Twitter
Sep 21
DIY Robocars via Twitter
RT @SahikaGenc: AWS DeepRacer & Hot Wheels Track https://youtu.be/4H0Ei07RdR4 via @YouTube
Sep 14
DIY Robocars via Twitter
Sep 8
DIY Robocars via Twitter
RT @davsca1: We are releasing the code of our Fisher Information Field, the first dedicated map for perception-aware planning that is >10x…
Sep 8
More…