Jules Mpano

RESEARCH

Research

Google Scholar

Self-Organizing Swarms & Robotics Lab

Designing Vision-Based Navigation Policies for Resource-Constrained Underwater Robots

Senior Thesis · Advised by Prof. Radhika Nagpal

Coral-reef monitoring needs autonomous platforms that can collect fine-grained data where human divers cannot persist. CoralBot — a fish-shaped robot with fin-based actuation, dual fisheye cameras, and an onboard Raspberry Pi 5 — is the hardware this thesis builds for. Getting it to navigate on its own means solving perception, control, and validation together.

The pipeline starts with a systematic evaluation of the Depth Anything V2 monocular depth model on underwater fisheye imagery. That depth signal feeds a six-region detection system that triggers obstacle avoidance, validated through physical pool deployments at two venues.

To go beyond reactive avoidance, I built a hardware-matched HoloOcean simulation and trained a twelve-model behavioral cloning sweep. Two ImageNet-pretrained depth policies reached 10/10 and 9/10 trajectory survival at 0.1% and 0.2% collision rates — with depth-channel mode collapse emerging as the binding constraint on vertical motion. A final COLMAP reconstruction benchmark characterizes why underwater scenes fail and proposes a SIFT-based feasibility diagnostic.

Depth Anything V2CoralBotRaspberry Pi 5HoloOceanBehavioral CloningCOLMAP

Princeton HCI Group

SorryIMissedThis

Advised by Prof. Parastoo Abtahi · CHI ’26 Extended Abstracts · ACM

Read the paper (DOI)

SorryIMissedThis is an AI-powered relationship-maintenance system that ingests conversation histories and generates context-aware prompts through a three-stage LLM pipeline: context and style modeling, template-guided synthesis, and a heuristic fallback. It’s privacy-preserving and local-first, sending only aggregated metadata off-device.

A counterbalanced within-subjects study (N=15) found history-informed prompts drove the engagement of generic ones, with significant gains in connectedness (r=0.82) and overall satisfaction (r=0.92).

FlaskReactAzure Cosmos DBLLM PipelineN=15 Study