[ ROBOTICS ENGINEER · AUTONOMOUS SYSTEMS & LEARNING-BASED CONTROL ] · TORONTO, ON

SOURABH
PRASAD

>
OPEN TO OPPORTUNITIES

I work where learning-based control meets real hardware: reinforcement learning for cross-embodiment generalization, multi-modal perception, and ROS2 software engineering — with the embedded and control experience to actually get robots running in the field.

EXPLORE TIMELINE
ROS / ROS2 Micro-ROS MoveIt2 PyTorch NVIDIA Isaac Sim Isaac Lab rsl_rl GStreamer MAVROS / MAVLink Docker Linux Git · CI/CD AWS C · C++ · Python MATLAB ESP32 · Teensy · Arduino SOLIDWORKS (CSWA) Deep RL Sensor Fusion
LOG · 06 ENTRIES · HOVER TO PREVIEW · CLICK TO PIN WORK ◄ │ ► EDU + PROJECTS
2026
2025
2023
2023
2022
2018
/DEEP DIVE

Deep dive

MEng Research · U of T · 2025

Cross-embodied quadruped navigation

Hierarchical navigation: a foundation-model visual planner driving a locomotion policy that transfers zero-shot to unseen robot morphologies.

Isaac Sim · PPO · NoMaD
71–100% success · 4 terrains
U of T · 2023

RGB-T sensor fusion

Late-fusion RGB + thermal detection pipeline for pedestrian detection in low light.

YOLOv5 · PyTorch · MLP rescoring
+28% average precision
Sentient Labs · 2021–2022

USV teleoperation stack

Fault-tolerant multi-channel teleop bridging ROS to ArduPilot over MQTT, P2P-VPN and redundant radio, with multi-camera streaming on AWS.

MAVROS · MAVLink · MQTT · GStreamer
<200 ms video latency
Sentient Labs · 2022

Grasping manipulator control

Low-level stepper control and kinematics under MoveIt2 planning, containerized with Docker for one-command deployment.

ROS2 · Micro-ROS · MoveIt2
80% autonomous grasp success
4 OF 4 PROJECTS SHOWN
/BLOG

Coming soon...

NOTIFY ME

Email me and I’ll ping you once when the first post lands. Nothing else.