1 / 15

Object Tracking Using Autonomous Quad Copter Carlos A. Muñoz

Object Tracking Using Autonomous Quad Copter Carlos A. Muñoz. Robotics, Intelligent Sensing & Control (RISC) Laboratory, School of Engineering, University of Bridgeport, 221 University Avenue, Bridgeport, CT 06604, USA. What is it, exactly?. A way to catch (tap) an object in the air.

dasan
Download Presentation

Object Tracking Using Autonomous Quad Copter Carlos A. Muñoz

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Object Tracking Using Autonomous Quad CopterCarlos A. Muñoz Robotics, Intelligent Sensing & Control (RISC) Laboratory,School of Engineering, University of Bridgeport,221 University Avenue, Bridgeport, CT 06604, USA.

  2. What is it, exactly? • A way to catch (tap) an object in the air. • Coordinates gather from Kinects • Automatically “sees” the object • Computes end destination • Sends signals to move the quad copter

  3. Hardware Used 3D Robotics Quad Copter Two Microsoft Kinects

  4. Overall Flow

  5. How does it know what the object is? • Object recognition Emgu CV: • Image converted to black and white • HSV image values • Slides determine final intensity threshold value • HSV values are added to the program • Red box shows detection of the object(s)

  6. HSV upper and lower limit selectors

  7. Adding object Object recognition EmguCV: Image converted to black and white HSV image values Slides determine final intensity threshold value Object is added to program Red shows detection

  8. How does it track it?Front camera Blue rectangle quad copter area Green is the minimum horizontal distance Green plus red is the maximum horizontal distance (1/8 ) Red lines are the vertical distances at different time intervals

  9. How does it compute the XY destination values? • Spot the object at a time interval [t] • Calculate initial horizontal velocity • Calculate initial vertical velocity Vx= x/t Vy = (y/t) – 0.5*g*t 2 Calculate different horizontal displacements until it reaches the copter. X = Vx * t and Y = Vy*t+0.5*g*t 2 • Use that same t to find the vertical displacement. • Determine new XY based on those values

  10. What about the Z value? • Bottom Kinect tracks two time intervals (uses same times ) • Calculates the final location using the distance formula dx = t2x – t1x dy = t2y – t1y Do the square of the distance between dx and dy: (dx2 + dy2) * k2 = tfinal (h) Solve for k then get the final point: x3 = x1 + t1x + dx*k y3 = x1 + t1y + dy*k

  11. How does it control the copter? • Camera tracks copter like an object • Virtual joystick created (Vjoy) • keyboard to joystick mapper configured • Program simulates keyboard presses • Once the new XYZ is acquired, execute the key presses

  12. .ini file Mission planner joystick configuration

  13. Does it really work? • Limitations: • Air resistance • Quad must be stabilized (minimal movement) • Quad is sensitive and very fast (crashed it 3 times) • Similar objects with color that is close to object

  14. Questions? • Website https://sites.google.com/site/carlosmrobotics/home • Email carlosm@my.bridgeport.edu

More Related