Sliding Autonomy Modrator Tool

This module is a plugin to the BYU HCMI lab's UAV Ground Control System (yet unnamed) interface that enables the UAV operator to manage UAV path planning autonomy using Sliding Autonomy. It provides real-time feedback to the user's what if path planning inquiries, so the user can choose the automatically generated path he/she approves and assign it to the UAV.

A simulation module is included to demonstrate how the tool works and also for performing user studies.

Plugin to Real Interface (C++)

Input/Output

Input Parameters:

  • Width, height, and matrix representing overall probability distribution map
  • (optionally) task-difficulty map

Output Parameters:

  • Width, height, and matrix representing probability distribution map
  • (optionally) task-difficulty map
  • Starting position in (x,y) coordinates
  • (Optionally) Ending position in (x,y) coordinates
  • Duration (in seconds)
  • UAV type: Fix-wing or Copter
  • Detection type: fixed amount (e.g., 5), fixed amount in percentage (e.g., 25% of original, takes 4 times to clear), fixed percentage (e.g., 50%, always 50% of the current amount)

Functions to be implemented

  • Plugin layer to existing interface
  • Request probability distribution from a repository
  • Update probability distribution in repository
  • Request task-difficulty map from a repository
  • Draw probably map using color map
  • User specify UAV parameters (UAV type, Detection type, etc.)
  • User specify search region (rectangular and can rotate)
  • Compute matrix representing search region probability map
  • User specify starting point (can drag and move, snap to grid), compute (x,y) coordinates wrt. matrix
  • User specify ending point (can drag and move, snap to grid), compute (x,y) coordinates wrt. matrix
  • Slider with minimum 1 maximum 3600 (seconds)
  • Slider maximum decreases as previous paths planned uses up the total 3600 seconds
  • As slider moves, periodically send request to IPPA to plan path (0.5 second intervals and 1 second duration change?)
  • Remember paths already planned to reduce computation and network traffic.
  • Draw path planned (as slider moves, remove previous path and draw new one)
  • Keep approved path waypoints
  • Use ending point from previous approved path as starting point for next segment
  • Button for user to approve path for current segment.
  • Field Trial
  • Write Field Robotics paper

Simulation Tool (C#)

Input Parameters:

  • Matrix representing probability distribution map
  • (optionally) task-difficulty map
  • Path returned from IPPA (a list of waypoints in (x,y) coordinates)

Output Parameters:

  • Width, height, and matrix representing probability distribution map
  • (optionally) task-difficulty map
  • Starting position in (x,y) coordinates
  • (Optionally) Ending position in (x,y) coordinates
  • Duration (in seconds)
  • UAV type: Fix-wing or Copter
  • Detection type: fixed amount (e.g., 5), fixed amount in percentage (e.g., 25% of original, takes 4 times to clear), fixed percentage (e.g., 50%, always 50% of the current amount)

Functions to be implemented

  • Load probability distribution map
  • Load task-difficulty map
  • Draw probably map using color map
  • User specify UAV parameters (UAV type, Detection type, etc.)
  • User specify starting point (can drag and move, snap to grid), compute (x,y) coordinates wrt. matrix
  • User specify ending point (can drag and move, snap to grid), compute (x,y) coordinates wrt. matrix
  • Slider with minimum 1 maximum 3600 (seconds)
  • Slider maximum decreases as previous paths planned uses up the total 3600 seconds
  • As slider moves, periodically send request to IPPA to plan path (0.5 second intervals and 1 second duration change?)
  • Remember paths already planned to reduce computation and network traffic.
  • Draw path planned (as slider moves, remove previous path and draw new one)
  • Keep approved path waypoints
  • Use ending point from previous approved path as starting point for next segment
  • Button for user to approve path for current segment.
  • Run user study and collect data
  • For user study
    • Log everything.
    • Select test maps (bi-modal with overlapping and complex)
    • Design training materials and training scenarios
    • Enable fixed flight patterns (lawnmower pattern, grid spiral pattern)
    • Count-down timer to inform user of how much time is left (total 5 minutes)
    • At the end show how many objects are collected.
    • Manually set waypoints (in grid)
    • Record task completion time.
    • Record number of mouse clicks.
    • Secondary task (recognizing audio signals of call signs).
    • Record secondary task performance.
    • Demonstrate fully automatically planned path and fly UAV accordingly.
  • Write HRI paper

Current To Do List

  • Manipulate terrain height map at run-time.
  • Color map terrain height map at run-time.
  • Export Unity 3D app and see.
  • Research how to include Unity in C# apps.
  • Load DistMap to memory
  • Load DiffMap to memory
  • Display DiffMap
  • Display DistMap
  • Vacuum as UAV flies
  • Allow manual flights
  • Allow flight patterns
  • Slider for resolution
  • Slider for flight duration
  • Network call to server
  • 100×100 cells for map
  • 24×24 cell size
  • 12m/s flight speed (and accelerate)
wisar/slidemod.txt · Last modified: 2014/08/11 19:48 by tmburdge
Back to top
CC Attribution-Share Alike 4.0 International
chimeric.de = chi`s home Valid CSS Driven by DokuWiki do yourself a favour and use a real browser - get firefox!! Recent changes RSS feed Valid XHTML 1.0