Autonomous navigation of miniaturized robots (e.g., nano/pico aerial vehicles) is currently a grand challenge in robotics research due to the need for processing a large amount of sensor data (e.g., camera frames) with limited onboard computational resources. Enabling efficient computing is critical for this task. Prof. Vivienne Sze, Prof. Sertac Karaman, and their team have developed several accelerated and low power computing modules for autonomous navigation, namely FastDepth (depth estimation), FSMI (Fast Shannon Mutual Information), and Navion (Visual-Inertial Odometry). As part of this research project, I aim to integrate these modules onto the MIT RACECAR to demonstrate real-time autonomous navigation.

“I am participating in SuperUROP to experience research after having taken relevant classes like 6.111 (Introductory Digital Systems Laboratory) and 6.115 (Microcomputer Project Laboratory). I hope to take away valuable skills in hardware design, as well as technical communication through the accompanying class 6.UAR.”