Discrete Collision Detection In High Performance Computing (GPU Acceleration System).
2
ABSTRACT
DISCRETE COLLISION DETECTION IN illGH PERFORMANCE
COMPUTING (GPU ACCELERA TJON SYSTEM)
(Keywords: 3D Visuali=ation. GPU. Collision Detection. Virtual Environment)
Virtual environment research has become an important research area in computer
graphics due to the need to visualise virtual environment as realistic as possible. Virtual
environment consists of elements such as culling, lighting, shadows, collision detection
and others that will determine the realism level of simulation. One of the key components
that need further consideration is to equip the synthetic world with fast and efficient
collision detection approach so that virtual environment can be done realistically. In
general , detecting object interference is an essential element in many three-dimensional
(3 D) application areas such as virtual reality, animation, simulation, game and robotics.
Virtual environment is quite useful today as it can represent any physical form of
engineering technology into 3D visualization . Thus, collision detection has become one
of the important elements that bring virtual world to imitate the realism of a real world
especially for medical simulation, pressure system, and many more. Hence, many issues
involving virtual env ironment and collision detection have been investigated. The
motivation is to come out with rea listic simulation when performing collision detection
for computer games and animation. The purpose of this research is to study the discrete
collision detection technique that can intens ify and boost up performance of collision
detection scheme in virtual environment. It is believed that the common collision
technique can be enhanced by using high performance computing system based on GPU
acceleration system in order to increase the magnitude of higher detection and faster
collision system. ln this research, our contribution concentration is on Bounding Volume
Hierarchies (BVH) technique using high performance computing based on GPU
acceleration in order to perform fast and efficient discrete collision detection to be used in
virtual environment.
Key Researchers:
Hamzah Asyrani Bin Sulairnan (Head)
Muhainmad Noorazlan Shah Bin Zainudin
Juwita Binti Mohd Sultan
Ahmad Sayuthi Bin Mohamad Shokri
Fakrulradzi Bin Idris
E-mail
Tel. No.
Vot No,
asyrani@utem.edu.my
06-5552150
PJP/20 IOIFKEKK(43E)/S00783
© Universiti Teknikal Malaysia Melaka
ABSTRACT
DISCRETE COLLISION DETECTION IN illGH PERFORMANCE
COMPUTING (GPU ACCELERA TJON SYSTEM)
(Keywords: 3D Visuali=ation. GPU. Collision Detection. Virtual Environment)
Virtual environment research has become an important research area in computer
graphics due to the need to visualise virtual environment as realistic as possible. Virtual
environment consists of elements such as culling, lighting, shadows, collision detection
and others that will determine the realism level of simulation. One of the key components
that need further consideration is to equip the synthetic world with fast and efficient
collision detection approach so that virtual environment can be done realistically. In
general , detecting object interference is an essential element in many three-dimensional
(3 D) application areas such as virtual reality, animation, simulation, game and robotics.
Virtual environment is quite useful today as it can represent any physical form of
engineering technology into 3D visualization . Thus, collision detection has become one
of the important elements that bring virtual world to imitate the realism of a real world
especially for medical simulation, pressure system, and many more. Hence, many issues
involving virtual env ironment and collision detection have been investigated. The
motivation is to come out with rea listic simulation when performing collision detection
for computer games and animation. The purpose of this research is to study the discrete
collision detection technique that can intens ify and boost up performance of collision
detection scheme in virtual environment. It is believed that the common collision
technique can be enhanced by using high performance computing system based on GPU
acceleration system in order to increase the magnitude of higher detection and faster
collision system. ln this research, our contribution concentration is on Bounding Volume
Hierarchies (BVH) technique using high performance computing based on GPU
acceleration in order to perform fast and efficient discrete collision detection to be used in
virtual environment.
Key Researchers:
Hamzah Asyrani Bin Sulairnan (Head)
Muhainmad Noorazlan Shah Bin Zainudin
Juwita Binti Mohd Sultan
Ahmad Sayuthi Bin Mohamad Shokri
Fakrulradzi Bin Idris
Tel. No.
Vot No,
asyrani@utem.edu.my
06-5552150
PJP/20 IOIFKEKK(43E)/S00783
© Universiti Teknikal Malaysia Melaka