Register now After registration you will be able to apply for this opportunity online.
Events and Frames Fusion for Visual Odometry on Nano-sized Drones
Autonomous nano-sized drones hold great potential for robotic applications, such as inspection in confined and cluttered environments. However, their small form factor imposes strict limitations on onboard computational power, memory, and sensor capabilities, posing significant challenges in achieving autonomous functionalities, such as robust and accurate state estimation. State-of-the-art (SoA) Visual Odometry (VO) algorithms leverage the fusion of traditional frame-based camera images with event-based data streams to achieve robust motion estimation. However, existing SoA VO models are still too compute/memory intensive to be integrated on the low-power processors of nano-drones. This thesis aims to optimize SoA deep learning-based VO algorithms and enable efficient execution on MicroController Units.
Autonomous nano-sized drones hold great potential for robotic applications, such as inspection in confined and cluttered environments. However, their small form factor imposes strict limitations on onboard computational power, memory, and sensor capabilities, posing significant challenges in achieving autonomous functionalities, such as robust and accurate state estimation. State-of-the-art (SoA) Visual Odometry (VO) algorithms leverage the fusion of traditional frame-based camera images with event-based data streams to achieve robust motion estimation. However, existing SoA VO models are still too compute/memory intensive to be integrated on the low-power processors of nano-drones. This thesis aims to optimize SoA deep learning-based VO algorithms and enable efficient execution on MicroController Units.
**Prerequisites**
Proficiency in Python and C programming. Background in Deep Learning. Experience in programming MicroControllers.
Autonomous nano-sized drones hold great potential for robotic applications, such as inspection in confined and cluttered environments. However, their small form factor imposes strict limitations on onboard computational power, memory, and sensor capabilities, posing significant challenges in achieving autonomous functionalities, such as robust and accurate state estimation. State-of-the-art (SoA) Visual Odometry (VO) algorithms leverage the fusion of traditional frame-based camera images with event-based data streams to achieve robust motion estimation. However, existing SoA VO models are still too compute/memory intensive to be integrated on the low-power processors of nano-drones. This thesis aims to optimize SoA deep learning-based VO algorithms and enable efficient execution on MicroController Units.
**Prerequisites** Proficiency in Python and C programming. Background in Deep Learning. Experience in programming MicroControllers.
This project aims to develop a deep neural network (DNN) for Visual Odometry that fuses traditional frame-based images with event camera data streams. The focus will be on optimizing the SoA model to reduce its computational complexity and memory footprint. The algorithm will be validated on real and/or simulated datasets to assess its robustness, and deployed on a microcontroller to profile its execution latency.
This project aims to develop a deep neural network (DNN) for Visual Odometry that fuses traditional frame-based images with event camera data streams. The focus will be on optimizing the SoA model to reduce its computational complexity and memory footprint. The algorithm will be validated on real and/or simulated datasets to assess its robustness, and deployed on a microcontroller to profile its execution latency.
Interested candidates should send their CV, transcripts (bachelor and master), and descriptions of relevant projects to Lorenzo Lamberti [llamberti (at) iis (dot) ee (dot) ethz (dot) ch], Marco Cannici [cannici (at) ifi (dot) uzh (dot) ch], Nico Messikommer [nmessi (at) ifi (dot) uzh (dot) ch], Luca Benini [lbenini (at) iis (dot) ee (dot) ethz (dot) ch], and Davide Scaramuzza [sdavide (at) ifi (dot) uzh (dot) ch].
Interested candidates should send their CV, transcripts (bachelor and master), and descriptions of relevant projects to Lorenzo Lamberti [llamberti (at) iis (dot) ee (dot) ethz (dot) ch], Marco Cannici [cannici (at) ifi (dot) uzh (dot) ch], Nico Messikommer [nmessi (at) ifi (dot) uzh (dot) ch], Luca Benini [lbenini (at) iis (dot) ee (dot) ethz (dot) ch], and Davide Scaramuzza [sdavide (at) ifi (dot) uzh (dot) ch].
