Sabeane Escobedo Palacios
University of California, Berkeley
Faculty Advisor: Prof. Dava Newman
Research Supervisor: Ireland Brown
Department: Aeronautics and Astronautics
Biography
Sabeane Escobedo is a rising junior at UC Berkeley, double-majoring in Aerospace
Engineering and Bioengineering. As a Mexican-Salvadoran student, she is committed to expanding
access and representation in STEM. Her interest in bioastronautics began after completing
underwater astronaut training, where she saw how aerospace engineering, biology, and medicine
could converge to protect human life in extreme environments. Sabeane has conducted research at
NASA Ames Research Center and the Berkeley Biomechanics Laboratory, and is active in Space
Technologies at Cal (STAC) and AIAA at Berkeley. This summer, she is working in Professor Dava
Newman’s Human Systems Lab at MIT, developing virtual/augmented reality navigation interfaces
for future lunar exploration with advanced spacesuit technologies. She hopes to pursue an MDPhD
in bioastronautics, designing systems that improve astronaut health and translate discoveries
from spaceflight to patient care on Earth. She brings curiosity, initiative, and interdisciplinary
commitment to research with meaningful human impact.
Development of a MATLAB-to-Unreal Engine Software-in-the-Loop Architecture
for Horizon-Based Lunar EVA Navigation Displays
Sabeane Escobedo Palacios1,2, Ireland M. Brown3, and Dava Newman3 4
1Department of Mechanical Engineering, University of California – Berkeley
2Department of Bioengineering, University of California – Berkeley
3Department of Aeronautics and Astronautics, Massachusetts Institute of Technology
4MIT Media Lab, Massachusetts Institute of Technology
Future lunar extravehicular activities require astronauts to maintain heading and position awareness
during traverses without continuous Global Navigation Satellite System coverage. This study builds
on Brown’s horizon-based heading estimation framework, which derives heading, uncertainty, and
terrain observability from monocular imagery and orbital digital elevation models. To translate
these outputs into an interface and validation platform, this project develops a MATLAB-to-Unreal
software-in-the-loop architecture. The architecture integrates heading, waypoint bearing, range
to goal, observability, navigation-aid state, and lateral uncertainty, while partitioning information
between a minimal helmet-mounted head-up display and a wrist display for expanded navigation,
biometrics, and location-dependent radiation-risk data. Three prototypes support precomputed
playback, bidirectional co-simulation, and closed-loop testing with Unreal-rendered imagery as
input. Metrics include data fidelity, latency, update rate, heading root-mean-square error against
Unreal ground truth, uncertainty calibration, match-failure rate, and final and cross-track navigation
error under lighting and observability conditions. Building on Brown’s Apollo 17 baseline, the
MATLAB-to-Unreal testbed evaluates whether navigation benefits persist and uncertainty can be
communicated when outputs are integrated into displays under timing, lighting, and observability
constraints, establishing a verification pathway toward augmented reality field testing.
