Haptic rendering method based on generalized penetration
depth computation
Yi Li
a
, Yin Zhang
a
, Xiuzi Ye
b,
n
, Sanyuan Zhang
a
a
Computer Science and Technique, Zhejiang University, 310027 Hangzhou, China
b
College of Mathematics & Information Science, Wenzhou University, 325003 Wenzhou, China
article info
Article history:
Received 2 October 2014
Received in revised form
4 December 2014
Accepted 5 December 2014
Available online 18 December 2014
Keywords:
Haptic rendering
6-DoF
Generalized penetration depth
Optimization
abstract
We present a new haptic iterative technology of 6-DoF haptic rendering based on the
generalized penetration computation method. Generalized penetration depth is the
minimum translational and rotational motion to separate two overlapping rigid objects;
we used an iterative, local optimization method to approximate generalized penetration
depth, and map the computation results to compute the virtual contact haptic feedback,
force and torque. We successfully integrated the algorithm into an off-the-shelf 6-DoF
haptic device (Phantom Premium). Our rendering algorithm can handle highly complex
polygon models and make no assumption about the underlying geometry and topology. In
order to resolve the instability problems, we use an optimization method to decrease the
subtle rotation motion and think about the feedback force direction in the physical
environment. The experimental result shows that our rendering method can generate
stable and realistic haptic feedback.
& 2014 Elsevier B.V. All rights reserved.
1. Introduction
Recently , man y graph-based models have been applied in
multimedia and computer vision. They can be used as
geometrical image descriptors [22] to enhance image cate-
gorizati on. F or insta nce, these methods can be used as image
high-order potential descrip tors of superpixels [23–25].The
graph-based descriptors can be used as general image
esthetic descriptors to improve image esthetics ranking,
photo retargeti ng and cr opping [26–29]. On the other hand,
people want to get the tactile feelings about the virtual
models. In the field of digital entertainment, medical training
[1–3], ph y sical simulation, etc., people think that haptic
rendering technology is an important research direction in
haptics. The user can touch the 3D object in virtual env iron-
ment and get a realistic tactile sensation; this w ould be
propitious to experience better interaction with the virtual
reality. Haptic rendering is the technique that can generate a
realistic force feedback to the user when the virtual models
collide with each other, accomplished by a haptic device, as
Phantom Premium and Haptic Virtuose. And the virtual
objects w ould not overlap in visualization. Ordinarily , we
hope to get a real phy sic al feedback and manipulat e the
virtual object neatly to achieve complicated tasks. Thus,
multiple degrees- of-freedom are very important in haptic
rendering, especially six-degrees-of-freedom (6-DoF) [4,5].
Three-degrees-of-freedom (3-DoF) haptic rendering method
which is a foundation technique has got broad and deep
study . Nevertheless, virtual torq ue is more necessary to get a
realistic haptic feedback; the 6-DoF haptic rendering method
that includes both force and torque feedback is extremely
challenging [6,19,7], most of all in high comple xity object
interact ion. Generally speaking, the 6-DoF haptic rendering
techniq ue can be classified into two catalogs [8,9]:penalty-
based hapti c r endering method and constraint-based haptic
rendering method. Being relative t o the constraint-based
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/sigpro
Signal Processing
http://dx.doi.org/10.1016/j.sigpro.2014.12.004
0165-1684/& 2014 Elsevier B.V. All rights reserved.
n
Corresponding author.
E-mail address: Yexiuzi@wzu.edu.cn (X. Ye).
Signal Processing 120 (2016) 714–720