New Optitrack System
New Optitrack system up and running at O134 inside netted space
Tracks rigid bodies and human bodies
Can stream data to ROS using vrpn_client
Contact Amir amir.moghaddam@qut.edu.au for booking
|
|
Internship Placement
This year QUT have added the STEM Internships to the Future You STEM Summit program, delivered in the mid-Semester break, 27 – 30 September. The year 11 and 12 students will participate in an internship placement OR STEM taster practicals from Faculties of Health, Science and Engineering. Simone Long (Young Accelerators Program Manager) is seeking help with securing research placements for students who chose the Internship stream of the program (72 students).
If researchers are interested in providing a placement, can they please contact Simone via stemschools@qut.edu.au, or on 0481 011 732.
|
|
|
QCR Promotion Material
Linktree is a great way to promote QCR social media without using multiple hyper links
|
|
|
PhD Awarded
Congrats Dr Valtteri Kallinen - Thesis completed 17 June 2022
Thesis title: “Collision Risk Modelling for Unmanned Aircraft Separation and Traffic Management”
Principal supervisor: Aaron McFadyen
Associate supervisors: Jason Ford
|
|
PhD Awarded
Congratulations to Dr Juan Sandino Mora who has been awarded a PhD for his thesis "Autonomous decision-making for UAVs operating under environmental and object detection uncertainty" supervised by Felipe Gonzalez, Frederic Maire, and external supervisors Dr Peter Caccetta and Dr Conrad Sanderson from Data61.
Juan is now working as a Research Fellow for the ARC Securing Antarctica Environmental Future(SAEF) program.
|
|
|
Journal Article
Title: An Anti-Attack Method for Emotion Categorization from Images
Applied Soft Computing Journal: Impact Factor 8.263, Shehu, Browne, Eisenbarth,
The hypothesis is that emotions displayed in facial images are more than patterns of pixels. Unlike other methods that have exhibited large performance decreases (up to 79%), the Facial features method was strongly resistant to all attacks, achieving high accuracy with only little (< 9.3%) or no decrease with different changes made to the images of CK+ and the KDEF databases.
____________________________________
|
|
Paper Accepted
Title: Predicting to Improve: Integrity Measures for Assessing Visual Localisation Performance
Helen Carson, Jason Ford, Michael Milford (IEEE RA-L Paper)
|
|
|
Paper Accepted
Gavin Suddrey, Ben Talbot and Frederic Maire have had their paper Learning and Executing Re-usable Behaviour Trees from Natural Language Instruction accepted into RAL/IROS 2022
|
|
Paper Accepted
Fabio Ruetz, Thierry Peynot, Niko Suenderhauf, Emili Hernandez (Emesent) and Paulo Borges (CSIRO) have had their paper “NDT-FTM: Normal Distributed Transform Forest Traversability Mapping” accepted into IROS 2022.
Fabio, as a co-author in collaboration with the mapping team at CSIRO, have had their paper “OHM: GPU Based Occupancy Map Generation” accepted IROS, RAL (pending).
|
|
|
Paper Accepted
Benjamin Moshirian and Pauline Pounds (UQ) have had their paper
“Rotor Array Synergies for Aerial Modular Reconfigurable Robots” accepted into IROS 2022.
|
|
Paper Accepted
Ben Burgess-Limerick, Christopher Lehnert, Jurgen Leitner, Peter Corke got their IROS Paper Accepted
“Eyes on the Prize: Improved Perception for Robust Dynamic Grasping”
|
|
|
Paper Accepted
Improving Worst Case Visual Localization Coverage with Dynamic Camera Selection
Stephen Hausler, Ming Xu, Sourav Garg, Punarjay Chakravarty (Ford Motor Corporation), Shubham Shrivastava (Ford), Ankit Vora (Ford), Michael J Milford (IEEE RA-L paper)
|
|
Paper Accepted
A RoboStack Tutorial: Using the Robot Operating System alongside the Conda and Jupyter Data Science Ecosystem
Tobias Fischer, Wolf Kristian Vollprecht, Silvio Traversaro, Sean Yen, Carlos Herrero, Michael J Milford (IEEE RAM paper)
|
|
|
IROS2022 Papers with QCR Collaborators Uni Essex
SwitchHit: A Probabilistic, Complementarity-Based Switching System for Improved Visual Place Recognition in Changing Environments Priors, Maria Waheed, Michael J Milford, Klaus McDonald-Maier, Shoaib Ehsan
Highly-Efficient Binary Neural Networks for Visual Place Recognition, Bruno Ferrarini, Michael J Milford, Klaus McDonald-Maier, Shoaib Ehsan (original stand-alone conference paper)
Binary Neural Networks for Memory-Efficient and Effective Visual Place Recognition in Changing Environments, Bruno Ferrarini, Michael J Milford, Klaus McDonald-Maier, Shoaib Ehsan (IEEE TRO paper)
|
|
RVSU Challenge Concluded
Robotic Vision Scene Understanding Challenge part of the Embodied AI workshop on 19 June.
23 registered participants across 7 countries
Over 100 submissions to the EvalAI servers
5 teams placed on the leaderboard
3 teams receiving GPU and/or cash prizes
Full challenge result info http://cvpr2022.roboticvisionchallenge.org/
Full workshop info https://embodied-ai.org/
|
|
|
|
Featured in ICVP Project Report
The project that Komol worked on with the Department of Transport and Main Road during his Master's study: the Ipswich Connected Vehicle Pilot Project (ICVP) has published the final report publicly. Komol has been acknowledged in the project report. He contributed to data cleaning, vehicle movement classification in intersections, identifying erroneous and missing trajectories due to GPS error, and predicting drivers turning behaviour and red light violation behaviour using AI technologies.
Report Link: https://lnkd.in/dgxfMZCx
Website Link: ipswich-connected-vehicle-pilot/
It was an interesting project on connected vehicle technology. The project includes a field operation test of 350 connected vehicles equipped with wireless communication devices and tested over a nine-month period at 29 different intersections on Ipswich Road, Queensland.
Komol also published a scientific research paper related to ICVP project during his Master's study period and the other two papers are under review. See the link to the published research
__________________________________________________________________________
|
|
Diversity
2022 CALD Women in Technology Program
WiT is one of the biggest communities supporting women in STEM and the champions of change who empower their success. They offer training and technology programs, such as those aimed for culturally & linguistically diverse (CALD) people. The faculty will pay the registration fee for this particular program - please contact Andrew Bradley.
>READ MORE
|
|
|
Leah Heiss Visit
Leah did a talk on Tuesday 12 July “Supercharging Creativity for Complex Challenges” in the Research Seminar Series
|
|
|
Coming Up
EResearch Showcase 2022 - mid September
Office of eResearch is planning an ‘eResearch Showcase’ in mid Sept and wanted to offer QCR staff and students the chance to be a part of it. See email below:
We’re inviting researchers we work with to come and showcase their projects and highlight which eResearch Services have helped enable your project.
It will be held in The Cube at Gardens Point campus, with projects displayed as ‘posters’ on the screens, allowing you to stand by the poster and talk through your project with people as they move through the event. There is also the option to present a short talk about your project from our main ‘stage’ – this part is optional.
HDR students are very welcome to be a part of it, our team supports HDR students and any researchers at QUT. As this is a showcase to highlight the services we provide with real-world examples, we only request that anyone presenting at the showcase has engaged eResearch services in some way to support their research project. Please contact Ilana in the first instance if you wish to be involved!
|
|
|
Coming Up
Masterclass: 'How to develop relationships to advance your research'
|
|
|
Past Events
Navigating into the Future Launch
|
|
|
Past Events
Full list of award winners from the QCR Retreat
>READ MORE
|
|
|
Past Events
QUT Robotics Club Droid Challenge Event
Stephen Hausler presented at the QUT Robotics Club Droid Challenge event on 15 July. His talk was titled "Autonomous Navigation: From Pixels to Pose". The talk was about localization for autonomous vehicles. In the first half of his talk, he spoke about some highlights from his PhD research in visual place recognition, such as Patch-NetVLAD, and explained that visual place recognition is commonly used as a backup system when GPS is unavailable. In the second half of his talk he spoke about vision-only six degree of freedom localization for autonomous vehicles, which is used for centimeter accurate localization on a roadway. He briefly explained his recently published RA-L paper in this space (which autonomously selects the optimal camera for localization - https://ieeexplore.ieee.org/document/9830877) and mentioned his planned future work.
|
|
|
Missed a seminar? Catch up here
Date: 28 June
Title: Navigating the Grant Landscape
Speaker: Ilana Bolingford
Download here
|
|
|
Date: 5 July
Title: Exploiting Spatial-Temporal Information for Robotic Vision in Agriculture
Speaker: Prof Chris McCool
Download here
|
|
|
|
Date: 12 July
Title: Supercharging Creativity for Complex Challenges
Speaker: Aspro Leah Hoiss
Download here
|
|
|
Date: 19 July
Title: Technical Talk: Introduction to the Indurad Radar Sensor
Speaker: Guenes Minareci
Download here
|
|
|
Welcome to Dr Dimity Miller
Dr Dimity Miller joins us as Lecturer and Chief Investigator.
Dimity completed her PhD on how to obtain uncertainty and robustness in deep learning for robotic vision in 2021.
She worked as a postdoctoral research fellow in a joint position with the Trusted Networks Lab at QUT, the CSIRO Robotics and Autonomous Systems group and the Data61 Machine Learning and Artificial Intelligence Future Science Platform.
>READ MORE
|
|
Welcome to Josh Buckley
Josh is doing work in collaboration with the law faculty on privacy for visual localization systems, having already done a undergraduate project in this general area. He'll be remote / with law faculty rather than physically in QCR.
|
|
|
Welcome to Natasha Goucher
Natasha (who has already doing her final year project with Tobias Fischer) is doing an exploratory RA position and will be looking at a few things to start - furthering the 360 degree acquisition rig prototype developed by Sarah Baldwin, including some background research on manufacturing, economies of scale) and using it in some initial VPR experimentation
|
|
Welcome to Quinlan Barthelme
New PhD Student - Quinlan Barthelme has started with us. His topic is: “Navigating through dense leafy green environments for trimming/harvesting purposes”
|
|
|
… with Will Browne
What do you do in the Centre and what are you currently working on?
My title is a bit of a mouthful: Prof and Chair in Manufacturing Robotics, where I wear three hats (well branded T-shirts anyway). My position is funded through CSIRO, ARM Hub (Advanced Robotics Manufacturing Hub) and QUT, with focus on industrial engagement.
Academically, my current focus is on abstraction, continual learning and eXplainable AI.
What’s on your desk?
Apart from day-to-day stuff, a couple of small mobile robots that one day will get bimanual grippers to sort out the stuff. I also have a few iterations of a stroke orthotic device, which are waiting for the right project/student to advance it.
How would you describe your work to a child?
I help robots and their internal computer systems interact with the world more intelligently.
Three words to describe your work.
Artificial cognitive systems
What’s your dream travel destination?
Antarctica as my wife, Lydia, wouldn't stop smiling throughout the trip regardless of the cold!
Please finish these sentences:
I’m currently reading
'A Path Towards Autonomous Machine Intelligence' by Yann LeCun (the last novel was 'A Wild Sheep Chase' by Murakami too many months back)
And currently wishing
apart from more hours in the day to do cool stuff, not much.
My last meal would be
tiramisu (Lydia's extra potent version)
Final question! Tea, coffee or no caffeine?
Hot chocolate!!
Who do you nominate for next month?
Tobi
|
|
|
Bayesian Quickest Change Detection of an Intruder in Acknowledgements for Private Remote State Estimation
JM Kennedy, JJ Ford, DE Quevedo - arXiv preprint arXiv:2207.08329, 2022
2207.08329.pdf (arxiv.org)
Manipulator Differential Kinematics
J Haviland, P Corke - arXiv preprint arXiv:2207.01796, 2022
https://arxiv.org/pdf/2207.01796.pdf
Manipulator Differential Kinematics Part I: Kinematics, Velocity and Applications
J Haviland, P Corke - arXiv e-prints, 2022
https://ui.adsabs.harvard.edu/abs/2022arXiv220701796H/abstract
Manipulator Differential Kinematics Part II: Acceleration and Advanced Applications
J Haviland, P Corke - arXiv e-prints, 2022
https://ui.adsabs.harvard.edu/abs/2022arXiv220701794H/abstract
Contemporary Remote Sensing Tools for Integrated Assessment and Conservation Planning of Ice-free Antarctica
LP Valerio, J Shaw, F Gonzalez, S Robinson, K Helmstedt – Preprint EarthArXiv- 2022
Contemporary Remote Sensing Tools for Integrated Assessment and Conservation Planning of Ice-free Antarctica
How Many Events Do You Need? Event-based Visual Place Recognition Using Sparse But Varying Pixels
T Fischer, M Milford - arXiv preprint arXiv:2206.13673, 2022
https://arxiv.org/pdf/2206.13673.pdf
Improving Worst Case Visual Localisation Coverage via Place-specific Sub-Selection in Multi-camera Systems
S Hausler, M Xu, S Garg, P Chakravarty, S Shrivastava, A Vora, M Milford - arXiv preprint arXiv …, 2022
https://arxiv.org/abs/2206.13883
A Novel Model of Interaction Dynamics between Legged Robots and Deformable Terrain
A Vanderkop, N Kottege, T Peynot, P Corke – 2022 International Conference on Robotics and Automation 2022
https://ieeexplore.ieee.org/abstract/document/9812351
When the Sun Goes Down: Repairing Photometric Losses for All-Day Depth Estimation
M Vankadari, S Golodetz, S Garg, S Shin, A Markham, N Trigoni - arXiv preprint arXiv …, 2022
https://arxiv.org/pdf/2206.13850.pdf
Transmission Power Policies for Energy-Efficient Wireless Control of Nonlinear Systems
V Varma, R Postoyan, D Quevedo, I Morarescu - IEEE Transactions on Automatic Control 2022
Scopus - Document details - Transmission Power Policies for Energy-Efficient Wireless Control of Nonlinear Systems
Visibility Maximisation Controller for Robotic Manipulation
K He, R Newbury, T Tran, J Haviland, B Burgess-Limerick, D Kulic, P Corke, A Cosgun
Scopus - Document details - Visibility Maximization Controller for Robotic Manipulation
|
|
|
|
|
