Shaliah Fricas :)

I build |

Computer Engineering student, McNair Research Fellow, and Robotics Field Engineer focused on human-centered robotics, artificial intelligence, and embedded systems. I use hands-on technical experience to create reliable, user-focused engineering solutions.

03 — Research missions

Research Missions

Active and Completed Missions

Active mission McNair Scholar · ICARUS Lab · UMass Boston

Sensor reliability in autonomous robot navigation

Research question How do sensor-based navigation systems in autonomous robots respond to manipulated or incorrect sensor input — and how can we make them more reliable?

Fully assembled TurtleBot3 Burger — LiDAR unit on top, Raspberry Pi camera in a pink 3D-printed mount, stacked waffle plates with Raspberry Pi 4 and OpenCR — Field unit · TurtleBot3 Burger · LiDAR + Camera + Raspberry Pi 4 *tap to enlarge*

This research investigates how sensor-based navigation systems in autonomous robots respond to manipulated, delayed, noisy, or otherwise incorrect sensor input. Using a TurtleBot3 Burger, the project evaluates how unreliable LiDAR, camera, IMU, and related sensor data affect mapping, localization, obstacle avoidance, and navigation.

The research compares normal and manipulated conditions, measures navigation performance, and explores methods such as filtering, sensor redundancy, and validation checks to improve reliability — toward safer assistive robotics.

Objective: Quantify navigation degradation under unreliable sensing; test defenses that restore it
Mentor · Lab: Prof. Ioannis Zografopoulos · ICARUS Lab
Program: McNair Scholars · Feb 2026 — Present
Status: In progress — baseline mapping & sensor-manipulation trials

Technologies & equipment

TurtleBot3 Burger
ROS 2
Ubuntu
LiDAR
Camera
IMU
Raspberry Pi
OpenCR
OPAL-RT
Python
Navigation & mapping tools

Research proposal Robotics data GitHub

Posters, demonstrations, and build videos will be logged here as the mission progresses.

Mission complete UCaN Lab · UMass Boston · MassURC 2026

Hybrid radio/optical indoor positioning

Research poster: Experimental Testing and Analysis of a Hybrid Radio/Optical System for Indoor Positioning — Tap to enlarge the poster

GPS fails indoors, where healthcare, warehouses, and navigation need it most. Our testbed fused RF fingerprinting (works without line-of-sight) with visible light positioning (high accuracy when visible) so localization stays reliable across environments.

0% Motion-classification accuracy at a 2 ms window

0% At a 20 ms window — shorter windows capture motion better

What I accomplished — the machine learning

Collected & analyzed wireless signal data in MATLAB across the SDR testbed
Built machine-learning models for localization and environment / signal classification
Contributed to experimental design, technical documentation & conference presentations (MassURC, AFRL SDR University Challenge)

▋

UCaN Lab progress

Team: Giovani DeOliveira, Shaliah Fricas, David Malone, Austin Trinh, Shabnam Azizi · Dr. Michael B. Rahaim. Partially supported by the NSF ERI Program (Grant No. 2347514).

Mission accepted UMass Robotics & Engineering Club

MassRobotics Competition

Role: Engineering Robotics Team | Robotics, Automation & Controls
Status: Preparing
Deployment: Spring 2027

Building with the club's competition team ahead of the 2027 MassRobotics event — this card unlocks as an achievement once the mission is complete.

Finalist — U.S. Air Force Research Laboratory SDR University Challenge

Our UCaN Lab team was selected as a finalist in the 2025–26 Software Defined Radio University Challenge hosted by AFRL. We built the test infrastructure for the hybrid indoor-localization work — and as finalists, we're heading to Dayton, Ohio to present to AFRL judges in person.

DeOliveira · Trinh · Malone · Fricas

UMass Boston · UCaN Lab

Top finalists → in-person round