Page 36 - ASME DSCC 2015 Program

Technical Program

Proportional sEMG Based Robotic Assistance in an Isolated Wrist Automatic Alignment of Motion Capturing System Markers to a Small
Movement differential drive robot’s position and heading angle
Invited session paper. DSCC2015-9979 Contributed regular paper. DSCC2015-9878
Edward Artz, Amy Blank, Marcia o’Malley, Rice University, Houston, TX, Dejan Milutinovic, UC Santa Cruz, Santa Cruz, CA, United States,
United States David W. Casbeer, AFRL Aerospace Systems Directorate, Wright-Patterson
AFB, OH, United States
As a first step towards integration of sEMG for intent detection in robotic
rehabilitation after incomplete SCI, we investigate a proportional control Marker-based motion capturing systems are effective in tracking a robot
scheme for the MAHI Exo-II that would provide assistance in completing only if the marker configuration is somehow aligned with the robot’s body.
movements of the upper limb. To overcome the manual alignment, in this paper we propose an automat-
ic procedure for a small differential drive ground robot. The procedure
ConTRIBuTED SESSIon requires that the robot travels along an arbitrary curved trajectory. The pa-
1-12-1 WP3 Tracking Control Systems rameters describing the relative position between the marker configuration
George Bellows E 4:00pm–6:00pm
and the robot’s body are estimated. Once the parameters are known, they
can be used to compute the robot’s center and its heading angle. The ef-
Session Chair: Manish Kumar, University of Toledo fectiveness of the procedure is illustrated by an experiment and the results
Session Co-Chair: Dejan Milutinovic, University of California, Santa Cruz showing that the accuracy of the robot center position exceeds the accuracy
Image Processing and localization for Detecting and Tracking Wildland of the motion capturing measurements of the markers’ positions.
fires Extended State observer Based Controller Design for the Green Bank
Contributed regular paper. DSCC2015-9839 Telescope Servo System
Anoop Sathyan, Kelly Cohen, University of Cincinnati, Cincinnati, OH, Contributed regular paper. DSCC2015-9677
United States, Manish Kumar, University of Toledo, Toledo, OH, United Trupti Ranka, Mario Garcia-Sanz, Case Western Reserve University,
States
Cleveland, OH, United States, John ford, National Radio Astronomy
Unmanned Aerial Vehicles (UAVs) are being used for a wide variety of Observatory, Green Bank, WV, United States
applications including detecting and tracking wildland fires. Using UAVs for The Green Bank Telescope is a large flexible structure, requiring rms track-
fire-fighting purposes reduces the human involvement for this high risk job. ing error less than 3 arcseconds against internal and external disturbanc-
Such a mission involves locating the wildland fire, tracking the direction of es. We design extended state observer (ESO) based controller in various
spread of the fire and searching for human presence in the region. This configurations to improve tracking performance and increase disturbance
paper investigates the algorithmic development for the use of UAVs to rejection. Through extensive simulation study, the response of ESO based
detect and track wildland fires. This would involve using the fuzzy toolbox controllers and legacy PID controller are compared using time and frequen-
in MATLAB along with MICRODEM, a software which provides the Digital cy domain responses. We show that the ESO based controller when imple-
Elevation Model (DEM) for the region. The objective of this research is to mented in both position and velocity loop can give significant improvement
accomplish the following: 1) use genetic fuzzy based image processing tools in tracking performance and better disturbance rejection without increase in
to identify fire from the video feed obtained from the camera attached to the controller output.
UAV in real time 2) look for human presence in the region and 3) estimate
Analysis of the Improvement in Accuracy using Multiple Gyros
the location of the fire based on the geological data available for the region.
Contributed regular paper. DSCC2015-9928
Appearance-Based outdoor localization using Group lASSo
Shuyang li, Yizhou Wang, Masayoshi Tomizuka, University of California,
Regression
Berkeley, CA, United States
Contributed regular paper. DSCC2015-9865
The possibility to improve the accuracy of angular velocity estimate by opti-
Huan Do, Jongeun Choi, Michigan State University, East Lansing, MI,
United States mally combining multiple individual micro electromechanical systems (MEMS)
gyroscopes is analyzed in this paper. Kalman filter (KF) is used to fuse the
This paper presents appearance-based localization for an omni-directional numerous measurements into a single improved rate estimate. Based on the
camera that builds on a combination of the group Least Absolute Shrinkage virtual gyro idea, a revised approach can give an unbiased estimator regard-
and Selection Operator (LASSO) and the extended Kalman filter (EKF). A his- less of the true angular rate signal. Numerical simulations are performed to
togram that represents the population of the Speeded-Up Robust Features demonstrate the desirable performance, reducing the bias drift and noise.
(SURF points) is computed for each image, the features of which are select-
ed via the group LASSO regression. The EKF takes the output of the LASSO
regression-based first localization as observations for the final localization.
The experimental results demonstrate the effectiveness of our approach.

36