Minisymposium

I will discuss ray tracing for astrophysics applications, with a focus on black hole imaging. Astrophysical black holes, i.e., those which astronomers do observe in a near and far Universe, are not always surrounded by vacuum but are often embedded in extremely hot swirl of plasma rotating almost at the speed of light. The physical conditions in such environment are not achievable in our Earth laboratories hence observations of plasma in a close vicinity of black holes enable studies of plethora of exotic phenomena such as particle acceleration to extreme energies, light bending in strong gravity, or gravitational energy conversion to other forms of energy. Ray tracing is used to simulate emission intensity in black hole accretion flows, creating high-resolution synthetic images. Parameter estimation of black holes based on experimental data requires comparison against a large number of these synthetic images, which is computationally intensive. By adapting a general relativistic ray-tracing code for GPU execution, significant speedups are obtained relative to CPU execution. Though high resolution images are necessary to observe fine features in the system, such as photon rings, features in black holes range over a wide range of scale, requiring a multiscale approach to generating images.

Monte Carlo (MC) particle transport simulations are integral to (1) the high-energy physics event reconstruction process, and (2) estimating radiological quantities within nuclear reactors. This method involves tracking individual particles through a computational geometry using a random walk technique. In practice, this requires interweaving ray tracing operations with computationally expensive physics calculations. The Oak Ridge Advanced Nested Geometry Engine (ORANGE) is a state-of-the-art computational geometry library shared by the Celeritas MC high-energy physics code and the Shift MC nuclear reactor simulation code. This presentation will focus on the ray tracing methods within ORANGE that facilitate efficient execution on the GPUs of leadership-class supercomputers. These methods include reordering operations to leverage single instruction, multiple threads (SIMT) parallelism, hierarchical acceleration structure algorithms, and algorithms that exploit the structured configurations of physical systems. Results from production-level simulations on the Frontier supercomputer demonstrate the effectiveness of these methods.

Ray tracing offers powerful capabilities for simulating wave propagation in diverse scientific fields, including acoustics. This presentation focuses on its application to in-air acoustic simulations, furthering our understanding of biological echolocation (e.g., bats) and advancing the development of in-air 3D sonar arrays, associated signal processing, and robotic applications. Our work in this area has evolved over time, initially using simpler 2D intersection calculations and progressing towards ray tracing methods capable of handling 3D scenes. Simulating the complex pulse-echo interactions of sound waves within detailed dynamic environments, involving multiple bounces and material-dependent reflections, presents significant computational challenges for which we will discuss our transition towards using hardware-accelerated ray tracing GPUs. This direction provides the ability to simulate larger, more complex scenes with moving objects and unique acoustic surface properties. We will discuss how key domain-specific challenges are tackled, such as modeling diffraction, alongside the continued need to optimize ray traversal and reflection calculations for achieving desired simulation fidelity efficiently on parallel hardware. This simulation framework facilitates rapid prototyping and testing of bio-inspired sonar sensors and their applications, enhances our understanding of complex animal behaviors, and provides a valuable tool bridging computational physics, robotics, and bioacoustics.

This panel discussion, featuring the three speakers, will provide an opportunity for extended audience questions and deeper engagement with the topics presented. It will also allow the speakers to interact with one another, compare methods, and discuss overlapping challenges across their fields.