Developer

Hi,

Ok, I was contacted today to see if its possible to build a quadcopter that could flight for at least 10 minutes with a small camera recording a flooded tunnel (it has artificial illumination) that has 4 x 3 meter and the bottom is water. 

It his feasible? They don't want to control the quadcopter, they want it to go inside this tunnel and go back after X minutes, to survey the tunnel.

Some computer vision maybe? Or small LIDAR? But how? How can I contact? If is there someone that can do this, I can hire you for this task.

ALK 

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

Join diydrones

Email me when people reply –

Replies

  • Developer

    Is the water relatively still? If so a floating device/boat would be much simpler then a quad. Flying a hover based craft like a multicopter inside a relatively narrow tunnel like this is going to be very difficult, especially with all the turbulence created by the propellers in the closed area. And regardless this job would require dedicated hardware and software. No off the shell solution that I know of would do this this automated.

    If the job has a budget for development I would look at getting a Hokuyo laser range finder with a >180deg beam and point it up into the ceiling so that you would scan the inside circumference of the tunnel and use that to position the copter accordingly.

  • Tunnel = No GPS, and I don't think the APM has enough CPU power to create a accurate map on its own just by sonar... and using this map for orientation. I would aim first for a much longer flight time frame as the drone will be extremly slow. Blades need cover and if there are things hanging inside the tube, you already go into hardly possible on a usual budget.

    We are working on a "Tracker-Drone" for Nature Conservation Botswana at the moment but after that, we have a Inhouse Sonar Drone as the next project. Working on a existing map is so much easier from a start, my opinion as the programmer of the projects.

     

    But as your tunnel is floaded... why not use a float with jet propulsion.. should be easier and you get much longer operating time and weight is not that much of an issue. But interesting project, will be curious to see and hear more of it.

     

    cheers

    Jay

  • Sonar would be the way to do this, but you would need a custom flight mode coded for the APM.

    Altitude hold with sonar keep a multi-rotor a fixed distance from the ground, you'd need a similar mode but in multiple axes to keep it a fixed distance from the roof and walls.

    If the people needing the survey done have deep pockets it could be a fun little task for someone who knows the APM code well.

This reply was deleted.

Activity