DrDonald Dansereau

Project Opportunities

Title: Low-latency robotic imaging for fast driving, flight and manipulation

Summary of opportunity:

This work will develop new low-latency visual perception systems, enabling unprecedented levels of robotic autonomy where response time is critical.

Opportunity synopsis:

Emerging technologies like bio-inspired dynamic vision sensors and low-latency computing architectures hint at a path to highly responsive robotic vision.In this project you will explore novel low-latency approaches to machine vision, enabling the next generation of fast robotic driving, flying, and manipulation.In scope are custom optics, electronics, algorithms, and computing architectures to deliver highly responsive, safe robots capable of operating at unprecedented speeds. Example approaches include information-driven active and adaptive imaging, generalised structured light, dynamic vision sensors (event cameras), light field video, custom LiDAR, multi-bucket sensors, single-photon avalanche diodes (SPADs), and low-latency machine learning and computing architectures.

Working within the Australian Centre for Field Robotics (ACFR), you will have access to the state-of-the-art robots, facilities, dedicated technical staff, and mentorship available through this world-class research centre. The ACFR undertakes significant field robotics programs in autonomous driving, flight, agriculture, and underwater survey, providing rich opportunities for deployment and validation of novel perception systems.

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

Project Opportunities

Title: Developing new imaging devices to help robots see and do

Summary of opportunity:

This work will develop specialised imaging devices to improve robotic perception in challenging conditions.

Opportunity synopsis:

In this project you will explore novel optical systems and algorithms that endow robots with new kinds of visual sensing. This approach has already yielded super-human perceptual capabilities like imaging around corners, recording a person's pulse from changes in skin colour during a heartbeat, and directly imaging a pulse of light as it propagates through a scene.There is an opportunity to draw on a broad range of techniques from the computational imaging and optics communities, including fabrication of custom optics and nano-fabrication of diffractive elements and metamaterials. Applications arise anywhere robots encounter perceptual challenges including all-weather autonomous driving, drone flight, underwater survey, human-robot interaction, and locomotion on challenging terrain.

Working within the Australian Centre for Field Robotics (ACFR), you will have access to the state-of-the-art robots, facilities, dedicated technical staff, and mentorship available through this world-class research centre. The ACFR undertakes significant field robotics programs in autonomous driving, flight, agriculture, and underwater survey, providing rich opportunities for deployment and validation of novel perception systems.

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

Project Opportunities

Title: 5D light field video processing for robust robotic vision

Summary of opportunity:

This work will develop algorithms that enable robots to benefit from the rich information captured by emerging light field video cameras.

Opportunity synopsis:

Light field cameras see a 4D superset of what normal 2D cameras see, simultaneously capturing light rays' positions and directions. This promises greater robustness in low light and through rain and fog, as well as natively capturing higher-order light transport effects like specularity and transparency.In this project you will develop the algorithms needed to make sense of these cameras in a robotics context, with an emphasis on real-time performance with light field video. There are opportunities to construct camera prototypes or to work with emerging commercial devices with embedded CPU, GPU, and FPGA, and to apply machine learning or conventional approaches to algorithmic development. Applications arise anywhere robots encounter perceptual challenges including all-weather autonomous driving, drone flight, underwater survey, human-robot interaction, and locomotion on challenging terrain.

Working within the Australian Centre for Field Robotics (ACFR), you will have access to the state-of-the-art robots, facilities, dedicated technical staff, and mentorship available through this world-class research centre. The ACFR undertakes significant field robotics programs in autonomous driving, flight, agriculture, and underwater survey, providing rich opportunities for deployment and validation of novel perception systems.

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

Project Opportunities

Title: Long-range robotic imaging in participating media

Summary of opportunity:

This project will push the limits of long-range vision in participating media including fog and water, expanding the range of conditions under which robots can be safely deployed. Applications include all-weather autonomous driving, drone flight, and underwater survey.

Opportunity synopsis:

Recent imaging advances emerging from the optics, computational imaging, and computer vision communities point the way forward for improving imaging through water, fog, rain, dust, and smoke. This project will jointly design optics and algorithms to deliver better perception in challenging, real-world robotic imaging scenarios.Potential approaches include long-range ghost imaging, active and adaptive imaging, burst photography, light field video, single-photon avalanche diodes (SPADs), dynamic vision sensors, multispectral sensing, generalised structured light, time of flight, and LiDAR. Employing machine learning to make sense of novel imaging modalities can form a substantial aspect of this project, as can advancing a fundamental theory of information-driven camera evaluation and design.

Working within the Australian Centre for Field Robotics (ACFR), you will have access to the state-of-the-art robots, facilities, dedicated technical staff, and mentorship available through this world-class research centre. The ACFR undertakes significant field robotics programs in autonomous driving, flight, agriculture, and underwater survey, providing rich opportunities for deployment and validation of novel perception systems.

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

Project Opportunities

Title: Using machine learning to give robots new kinds of visual sensing

Summary of opportunity:

This work will develop the machine learning techniques and principles needed for robots to automatically interpret and benefit from a broad range of emerging imaging technologies.

Opportunity synopsis:

Recent imaging advances have yielded super-human perceptual capabilities like imaging around corners, recording a person's pulse from changes in skin colour during a heartbeat, and directly imaging a pulse of light as it propagates through a scene. These technologies do not see the world the same way a conventional camera does, and making use of them in robotics raises important new challenges.In this project you will expand current ideas in machine learning to bridge a gap between robotics and an expanding array of exciting new imaging technologies. Potential approaches include unsupervised and semi-supervised learning, active autonomous data collection, online learning, and new neural processing elements and architectures. Imaging technologies might include solid-state LiDAR, single-photon sensors, transient/femtosecond imaging, light field imaging, imaging around corners, and event-based dynamic vision sensors. Applications arise anywhere robots encounter perceptual challenges including all-weather autonomous driving, drone flight, underwater survey, human-robot interaction, and locomotion on challenging terrain.

Working within the Australian Centre for Field Robotics (ACFR), you will have access to the state-of-the-art robots, facilities, dedicated technical staff, and mentorship available through this world-class research centre. The ACFR undertakes significant field robotics programs in autonomous driving, flight, agriculture, and underwater survey, providing rich opportunities for deployment and validation of novel perception systems.

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

Project Opportunities

Title: Whole-system robotic perception: From photons to actions

Summary of opportunity:

This work will build two-way information flows between the imaging, mapping, and control systems of a robot, allowing operation in previously inaccessible dynamic and interactive spaces.

Opportunity synopsis:

Modern machine vision systems generally use one-way information flows to progressively compress visual reality down to the most salient information. In mammalian brains things are much more complex, with substantial information flows from higher-order systems back towards low-level vision.In this project you will introduce new information loops between low-level sensing, high-level perception, and control. This can include priming neural architectures and driving active sensing systems with high-level context. Fundamental questions arise in representing knowledge, uncertainty, and intentionality inside and outside the robot. The project can also explore the impact of these new information flows on system design, with new possibilities in embodied intelligence and joint design of whole robotic systems. Application areas arise where action is tightly linked to perception including grasping and manipulation, robotic surgery, drone flight in cluttered environments, and autonomous driving in crowded spaces.

Working within the Australian Centre for Field Robotics (ACFR), you will have access to the state-of-the-art robots, facilities, dedicated technical staff, and mentorship available through this world-class research centre. The ACFR undertakes significant field robotics programs in autonomous driving, flight, agriculture, and underwater survey, providing rich opportunities for deployment and validation of novel perception systems.