1
Autonomous Object Retrieval with a Robotic Arm Samantha Luber, Zach Oligschlaeger, John Scheible Abstract Through incorporating various object detection algorithms on images from an overhead camera, we are able to efficiently track balls and blocks in the robotic arm’s search space. By applying a planning algorithm, the robotic arm is able to retrieve reachable objects and move them to the collection bin. Ball Detection Optimized template matching algorithm scans through images from overhead camera Searches for balls using a stored yellow ball image as the template Block Detection Uses a Union Find algorithm to build up pixel blobs of the same color Determines blobs within a color and size threshold to be blocks Calculates block orientation with respect to the arm Planning Detects reachable objects Calculates movement required by each joint needed to reach object and plans joint movement to avoid collisions Detects if the object was missed and re-attempts retrieval if necessary Moves object to static goal location Arm Communication LCM message passing by subscribing and posting movement commands to each servo’s channel

Object-retrieving Autonomous Robotic Arm

Embed Size (px)

DESCRIPTION

Poster for our presentation on our Autonomous Robotic Arm that detects objects with an overhead camera, uses motion planning and reverse kinematics to retrieve objects, and places the objects in a bin. This project was done as part of the EECS 498 Autonomous Robotics Lab at the University of Michigan.

Citation preview

Page 1: Object-retrieving Autonomous Robotic Arm

Autonomous Object Retrieval with a Robotic Arm!!

Samantha Luber, Zach Oligschlaeger, John Scheible!

Abstract!Through incorporating various object detection algorithms on images from an overhead camera, we are able to efficiently track balls and blocks in the robotic arm’s search space. By applying a planning algorithm, the robotic arm is able to retrieve reachable objects and move them to the collection bin. !

Ball Detection!•  Optimized template matching

algorithm scans through images from overhead camera!

•  Searches for balls using a stored yellow ball image as the template!

Block Detection!•  Uses a Union Find algorithm

to build up pixel blobs of the same color!

•  Determines blobs within a color and size threshold to be blocks!

•  Calculates block orientation with respect to the arm!

Planning!•  Detects reachable objects!•  Calculates movement

required by each joint needed to reach object and plans joint movement to avoid collisions!

•  Detects if the object was missed and re-attempts retrieval if necessary!

•  Moves object to static goal location!

Arm Communication!•  LCM message passing by subscribing and

posting movement commands to each servo’s channel!