Robot arm pose estimation by pixel-wise regression of joint angles
Authors:
- 1 May 2016
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 12, 616-623
Abstract
To achieve accurate vision-based control with a robotic arm, a good hand-eye coordination is required. However, knowing the current configuration of the arm can be very difficult due to noisy readings from joint encoders or an inaccurate hand-eye calibration. We propose an approach for robot arm pose estimation that uses depth images of the arm as input to directly estimate angular joint positions. This is a frame-by-frame method which does not rely on good initialisation of the solution from the previous frames or knowledge from the joint encoders. For estimation, we employ a random regression forest which is trained on synthetically generated data. We compare different training objectives of the forest and also analyse the influence of prior segmentation of the arms on accuracy. We show that this approach improves previous work both in terms of computational complexity and accuracy. Despite being trained on synthetic data only, we demonstrate that the estimation also works on real depth images.Keywords
This publication has 24 references indexed in Scilit:
- Pose Estimation of Kinematic Chain Instances via Object Coordinate Regression2015
- Keypoint Recognition Using Random Forests and Random Ferns2013
- Hough-Based Tracking of Deformable Objects2013
- Class-Specific Hough Forests for Object Detection2013
- Decision Forests for Computer Vision and Medical Image Analysis2013
- Generalizing grasps across partly similar objects2012
- Automatic Localization of Interest Points in Zebrafish Images with Tree-Based Methods2011
- BlenSor: Blender Sensor Simulation Toolbox2011
- Quality assessment of dimensionality reduction: Rank-based criteriaNeurocomputing, 2009
- Energy-minimizing splines in manifoldsACM Transactions on Graphics, 2004