MRSD Project Progress Review
"Team HARP at Carnegie Mellon University made significant advancements in their MRSD project, including acquiring a shop vac for the suction system, finalizing PCB design, and developing perception algorithms for object detection. They also tested Inverse Kinematics actions on the PR2 arm and utilized the Smach state controller for motion planning. The team's goals for the next week involve finalizing hardware, procuring a computer for Kinect2, and demonstrating arm movements in a simulated environment."
Download Presentation
Please find below an Image/Link to download the presentation.
The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author. If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.
You are allowed to download the files provided on this website for personal or commercial use, subject to the condition that they are used lawfully. All files are the property of their respective owners.
The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author.
E N D
Presentation Transcript
Progress Review - 3 MRSD Project - 1 11/12/2015 Team HARP (Abhishek, Alex, Feroze, Lekha, Rick) Carnegie Mellon University
Suction Acquired shop vac for final suction system. Re-iterated on the prototype suction gripper design. Added the custom gripper in simulation.
Suction Finalized PCB design Designed and ordered components for electronics enclosure
Perception Developed algorithm to remove shelf from point cloud Improved clustering by switching from color-based to euclidian Estimated object based on bounding box dimensions compared to ground truth Compiled all vision subsystems into one complete pipeline
Perception Ran 1000 image test on point clouds from Rutgers database Accurately identified grasp position within 3 cm on over 56% of attempts Investigating results to identify ways to improve algorithm (70% required by FVE)
Platform Move actual PR2 arm using Inverse Kinematics action The Inverse Kinematics arm motion in simulator was tested on the actual robot The program was loaded to the robot s computer and executed.
Platform Smach state controller was used to call MoveIt planner The target pose was defined using the movable interactive marker inside rviz The arm was moved to target pose successfully Eventually, the perception algorithm will return the target pose to smach which will then be used to plan and execute arm motion
Goals for next week: Suction Final PCB and enclosure. Create test circuit with components and build final hardware. Perception Procure computer for Kinect2. Run algorithms directly grabbing real-time data from Kinect-2. Platform Use moveit_pr2 to avoid collisions. Demonstrate a simulation that can move the arms to an item, move base and move the arm to order bin.
