DrViorela Ila

Project Opportunities

Title: State Estimation and Planning in Dynamic Environments

Research Area:

Robotics, Computer Vision, Optimization

Opportunity synopsis:

In robotics, the conventional solutions for mapping dynamic environments focus on either trying to remove the data associated with moving objects, or detect and track the moving objects separately using traditional object tracking approaches. Ignoring dynamics in applications where the environment is cluttered with moving objects can drastically affect the safety of operations in such environments. Separately tracking moving objects, on the other hand, will produce representations of the world where the static and moving parts of the scene are decoupled. This can drastically affect the accuracy of the information used by the navigation and path planning algorithms, and in consequence the safe operation of autonomous vehicles.

The aim of this project is to develop an integrated framework to estimate and generate virtual representations of dynamic scenes that can be directly used by autonomous vehicles and intelligent systems to plan and perform tasks in a safe manner. https://robotics.sydney.edu.au/our-research/robotic-perception/

Successful candidates will have:

• A bachelor's degree in a relevant discipline
• Good mathematics and programming background.

How to Apply: To apply, please email viorela.ila@sydney.edu.au, with the subject line "PhD Application” and your name. Include the following:

• CV and cover letter
• Transcripts (can be unofficial)

--------------------------------------------------

Project Opportunities

Title: Robotic Perception Lab Scholarships

Research Area:

Sensing, mapping and planning in dynamic environments.

Opportunity synopsis:

The projects aim to develop and deploy on real robots theoretically well-founded solutions to robot localization, mapping and planning in dynamic environments.

Depending on interest and ability, candidates will investigate one or more of:

• Robust techniques for motion segmentation and tracking
• Novel spatial-temporal representations of dynamic environments
• Robust estimation and reconstruction of the dynamic scene
• Heterogeneous sensing for dynamic environments

Research Environment:

Embedded in the Australian Centre for Field Robotics, the Robotic Perception Lab is focused on delivering fundamental methodologies to sensing, mapping and planning in real-word dynamic environments.

The ACFR offers specialised labs and facilities, robotic platforms (underwater, agriculture, electric vehicles) and robotic field labs on-campus and in nearby off-campus sites. You will have access to mechanical and electronics workshops and a pool of technical staff to help realise your research ambitions. The University of Sydney offers a rich academic setting in a world class city, and the ACFR has strong ties to a network of nearby and international academic and industrial collaborators.

Successful candidates will have:

• A bachelor's degree in a relevant discipline
• Interest in robotics
• Strong background in mathematics and programming
• Excellent communication and interpersonal skills
• Hands-on experience with robotic platforms, ROS, Python, C++, and/or deep learning frameworks would also be an asset
• Creativity, curiosity, rigour and passion

How to Apply: To apply, please email viorela.ila@sydney.edu.au, with the subject line "PhD Application” and your name. Include the following:

• CV and cover letter and transcripts (can be unofficial)
• A link to a 2 minute video covering your strongest engineering, mathematical, programming skills and what you enjoy most about research.

----------Current Research Projects----------

Student Projects:

• Vision-based Motion Detection Using Deep-Learning Techniques
• Underwater Localization Based on 4 GPS Buoys
• Active 3D Reconstruction Using the UR5e Robotic Arm
• 3D Point Cloud Segmentation of Small Objects
• Construction and Test of a Low-Cost UAV for Visual Mapping
• Real-time 3D Reconstruction
• Multirobot Collaborative Scene Representation
• Transfer Learning Instance Level Semantic Segmentation
• Outdoor Scene Reconstruction with AR Headset
• Terrain Characterization Based on IMU Sensing