Australasian Conference on Robotics and Automation 2017

December 11-13, 2017 at The University of Technology Sydney

Each presentation is 12mins prez + 3mins Q&A (Total 15mins)

Registration Desk:
8:00 am-11:00 am: Monday

Monday, 11 December 2017
09.15 – 09.30: Welcome
09.30 – 10.00: Invited Talk: Ken Waldron – A Memoir of Fifty Years of Robotics Research
10.00 – 11.00: Session 1: Modelling and Control
11.00 – 11.30: Morning tea
11.30 – 12.30: Session 2: Student Award Nominee Papers
12.30 – 14.00: Lunch
14.00 – 15.30: Session 3: Human Robot Interaction and Systems
15.30 – 16.00: Afternoon tea
16.00 – 17.30: Session 4: Machine Learning
18.00 – 20.00: Opening Night Pizza/Drinks

Tuesday, 12 December 2017
09.00 – 10.30: Session 5: SLAM
10.30 – 11.00: Morning tea
11.00 – 12.45: Session 6: Perception
12.45 – 14.00: Lunch
14.00 – 15.30: Session 7: Robotic vision
15.30 – 16.00: Afternoon tea
16.00 – 17.30: Session 8: Path Planning
18.30 – 22.30: Banquet (including AGM)

Wednesday, 13 December 2017
09.00 – 10.30: Session 9: Sensing and Control
10.30 – 11.00: Morning tea
11.00 – 12.30: Australian Robotics and Vision Roadmap
12.30 – 13.30: Lunch
13.30 – 15.30: Tours (FEIT and Data Arena)

Invited Talk – A Memoir of Fifty Years of Robotics Research

Kenneth J Waldron Professor, Faculty of Engineering and Information Technology, U. of Technology, Sydney

Chair: Sarath Kodagoda

Dr. Waldron obtained the degrees of Bachelor of Engineering and Master of Engineering Science from the University of Sydney (Australia). He received his PhD from Stanford University in 1969. He was also awarded the degree of Doctor of Engineering by the University of Sydney
in 1999. After a period with Australian Iron and Steel Pty. Ltd.  He has held teaching appointments at Stanford, the University of New South Wales, and the University of Houston, before joining The Ohio State University in 1979. He served as department chairman from April 1st, 1993 to June 30th, 2000. He joined Stanford University on September 1st, 2000. He became Professor (Research) Emeritus of the Department of Mechanical Engineering, Stanford University in 2011. He holds a half-time appointment at the University of Technology, Sydney effective July 2007.
Dr. Waldron’s research interests include machine design in general, and particularly geometric, mechanical and software design of robots and computer coordinated mechanical systems. He has a particular interest in locomotory biomechanics and in the design of computer coordinated vehicles and robotic systems for use in unstructured environments. Dr. Waldron is a Fellow of The American Society of Mechanical Engineers and a Senior Member of IEEE. He was Technical Leader of the ASME Systems and Design Group and was President of IFToMM the International Federation for Promotion of Mechanism and Machine Science from 2000 to 2007. He received the ASME Leonardo da Vinci Award in March 1988, the Mechanisms Award in 1990, the Machine Design Award in 1994, the Ruth and Joel Spira Outstanding Design Educator Award in 2002 and the Dedicated Service Award in 2004. He was a speaker in the ASME Distinguished Lecturers Program from 1996 to 1999. He also received the Joseph F. Engelberger Award of the Robotic Industries Association in 1997.

Session 1: Modelling and Control
Chair: Jonathan Roberts

Steep Terrain Ascension Controller for Hexapod Robots Alberto Elfes (CSIRO), Navinda Kottege (CSIRO),
Thomas Molnar (CSIRO, University of Queensland), Ryan Steindl (CSIRO), Fletcher Talbot (CSIRO)
Active Task Design in Adaptive Control of Redundant Robotic Systems Wenjie Lu (University of Technology Sydney), Dikai Liu (University of Technology Sydney) digest
A New Manipulability Measure for the Control of CRAM: A Cable-driven Remote Access Manipulator  Ali Haydar Goktogan (Australian Centre for Field Robotics), Wilhelm Johan Marais (The University of Sydney) digest
 Loop Shaping Design Procedure for Quadrotor
Control with Weights Designed by Resolving a Constrained Non-linear
Optimization Problem
 Jonathan Currie (Auckland University of Technology), Joseph Thomas (Auckland University of Technology), David Wilson (Auckland University of Technology) paper

Session 2: Student Award(s) Nominee Papers
Chair: Bruce MacDonald

Quasi-Static Balance of a Bioinspired Robotic-Seagull Ornithopter Perching on a Wire Yuxin Pan (The University of Sydney), Ali Haydar Goktogan (Australian Centre for Field Robotics) paper
Energy-Optimal Kinodynamic Planning for Underwater Gliders in Flow Fields James Ju Heon Lee (University of Technology Sydney), Chanyeol Yoo (University of Technology Sydney), Raewyn Hall (DSG), Stuart Anstee  (DSG), Robert Fitch (University of Technology Sydney) paper
Simultaneous Optical Flow and Segmentation (SOFAS) using Dynamic Vision Sensor Timo Stoffregen (Monash University), Lindsay Kleeman (Monash University) paper
Modular Deep Q Networks for Sim-to-real Transfer of Visuo-motor Policies Fangyi Zhang (Queensland University of
Technology), Jürgen Leitner (Queensland University of Technology), Michael Milford (Queensland University of Technology), Peter Corke (Queensland University of Technology)

Session 3: Human Robot Interaction and Systems
Chair: Dikai Liu

 Augmented Telepresence for Remote Inspection with Legged Robots  Benjamin Tam (CSIRO), Navinda Kottege (CSIRO), Branislav Kusy (CSIRO) paper
 UoA Robot Manager Framework V2  Seung Ho Lee( The University of Auckland), Kijong Han (The University of Auckland), Min Ho Lee (The
University of Auckland), Lukas Vogel (Augsburg University), Jong Yoon Lim (The University of Auckland), Bruce A. MacDonald (The University of Auckland), Ho Seok Ahn (The University of Auckland)
 Effect of External Force and Bimanual Operation on Upper Limb Pose during Human-Robot Collaboration  Richardo Khonasty (University of Technology Sydney), Marc Carmichael(University of Technology Sydney), Dikai Liu (University of Technology Sydney), Stefano Aldini (University of Technology Sydney) digest
 Heart Simulator: A Periodic Pump to Simulate the Cardiac Motion in an Aortic Test-rig Siyuan Chen (University of Technology,
Sydney), Viet Hung Doan (University of Technology,
Sydney), Liang Zhao (University of Technology, Sydney),
 Human User Impressions of Damping Methods for Singularity Handling in Human-Robot Collaboration  Marc G. Carmichael (University of Technology
Sydney), Stefano Aldini (University of Technology Sydney), Dikai Liu (University of Technology Sydney)

Session 4: Machine Learning
Chair: Teresa Vidal Calleja

Learning to Navigate by Growing Deep Networks Thushan Ganegedara (The University of Sydney), Lionel Ott (The University of Sydney), Fabio Ramos (The University of Sydney) paper
Auxiliary Tasks To Improve Trip Hazard Affordance Detection On Construction Sites Sean McMahon (Queensland University of Technology), Tong Shen (The University of Adelaide), Niko Sünderhauf (Queensland
University of Technology), Ian Reid (The
University of Adelaide), Chunhua Shen (The University of Adelaide), Michael  Milford (Queensland University of
Row Following in Pergola Structured Orchards by a Monocular Camera Using a Fully Convolutional Neural Network Jamie Bell (University of Auckland), Bruce A. MacDonald (University of Auckland), Ho Seok Ahn (University of Auckland) digest

Session 5: SLAM
Chair: Jose Guivant

Efficient Active SLAM based on Submap Joining


Yongbo Chen (University of Technology,
Sydney), Shoudong Huang (University of Technology,
Sydney), Robert Fitch (University of Technology,
Sydney), Jianqiao Yu (Beijing Institute of Technology),
Sparse Point-Plane SLAM


Mehdi Hosseinzadeh (The University of
Adelaide),  Yasir Latif (The University of Adelaide), Ian
Reid (The University of Adelaide)
Robust Dense Optical Flow with Uncertainty for Monocular Pose-Graph SLAM Yonhon Ng (The Australian National University),
Jonghyuk Kim (The Australian National University), Hongdong Li (The Australian National University)
Vector Distance Function Based Map Representation for Robot Localisation Janindu Arukgoda (University of Technology,
Sydney), Ravindra Ranasinghe (University of Technology, Sydney), Lakshitha Dantanarayana (University of Technology,
Sydney), Gamini Dissanayake (University of Technology, Sydney), Tomonari Furukawa (Virginia Polytechnic Institude and State University)

Session 6: Perception
Chair: Alen Alempijevic

Motion States Inference through 3D Shoulder Gait Analysis and Hierarchical Hidden Markov Models  Julien Collart (University of Technology,
Sydney),  Robert Fitch (University of Technology,
Sydney), Alen Alempijevic (University of Technology, Sydney)
Using Planar Point Correspondence to Calibrate Camera Arrays for Light Field Acquisition Ashley W. Stewart (Queensland University of Technology), Donald G. Dansereau (Stanford University) paper
Fusing Lidar and Semantic Image Information in Octree Maps Julie Stephany Berrio Perez (The University of
Sydney),  James Robert Ward (The University of
Sydney), Stewart Worrall (The University of Sydney), Wei
Zhou (The University of Sydney), Eduardo Nebot (The
University of Sydney)

Session 7: Robotic Vision
Chair: Michael Milford

Extending Parallax Parameterised Bundle Adjustment to Stereo


Brenton Leighton (University of Technology,
Sydney), Liang Zhao (University of Technology, Sydney), Shoudong Huang (University of Technology, Sydney), Gamini Dissanayake (University of Technology, Sydney)
Straightening Sequence-Search for Appearance-Invariant Place Recognition using Robust Motion Estimation Sourav Garg (Queensland University of  Technology), Michael Milford (Queensland University of Technology), digest
Image Rejection and Match Verification to Improve Surface-Based Localization


James Mount (Queensland University of Technology),Michael Milford (Queensland University of Technology) digest
Enhancing Underground Visual Place Recognition with Shannon Entropy Saliency


Fan Zeng (Queensland University of Technology),Adam Jacobson (Queensland University of Technology), David Smith (Caterpillar), Nigel
Boswell (Caterpillar), Thierry Peynot (Queensland University of Technology), Michael J. Milford(Queensland University of Technology)

Session 8: Path Planning
Chair: Robert Fitch

Path planning for autonomous bulldozers


Masami Hirayama (Komatsu Ltd.), Jose Guivant
(University of New South Wales), Jayantha Katupitiya(University of New South Wales), Mark Whitty(University of New South Wales),
 A Distributed Search Algorithm for Multiple Robots with Collision Avoidance Xiaotian Yang (The University of New South Wales) paper
 The collision avoidance control algorithm of the UAV formation flight Jialong Zhang (Northwest Polytechnical Univerity) paper
 Insect Level Intelligence is Sufficient to Move Furniture Bimal Prakash Sharma (Queensland University of Technology), Frederic Maire (Queensland University of Technology) digest

Session 9: Sensing and Control
Chair: Mark Whitty

 Contactless Position Sensing and Control of Pneumatic Cylinders using a Hall Effect Sensor Array Tim Hojnik (CSIRO), Paul Flick (CSIRO), Jonathan Roberts (Queensland University of Technology) digest
 Optimal Sensing Geometry for Pseudorange and Bearing-Elevation Observations Jonghyuk Kim (The Australian National University), Weikun Chen (The Australian National University) paper
Design and Control of CRAM: A Highly Articulated Cable-driven Remote Access Manipulator for Confined Space Inspection Wilhelm Johan Marais (The University of Sydney), Ali Haydar Goktogan (The University of Sydney) paper
 Robust Heterogeneous Multi-Robot Task Allocation for Low-Intelligence Agents Nick Sullivan (The University of
Adelaide), Steven Grainger (The University of
Adelaide), Ben Cazzolato (The University of Adelaide)
A Survey on Inverse Dynamics Solvers for Cable-Driven Parallel Robots Yin Pok Chan (The Chinese University of Hong
Kong),  Jonathan Eden (The University of Melbourne), Darwin La (The Chinese University of Hong Kong), Denny Oetomo (The University of Melbourne)

Session 10: A National Roadmap for Robotics and Computer Vision in Australia
Organizer: Dr Sue Keay, COO, Australian Centre for Robotic Vision


Robotic Vision, the application of computer vision to robotics, enables robotic technologies to add value to industries that require automation in complex physical environments. The development of robotic vision technologies will lead to the creation of new companies, new jobs and will address a range of issues of national importance including a aging population, servicing remote communities, and dealing with labour shortages.

A Robotics and Computer Vision Roadmap for Australia is needed to demonstrate existing capability and future possibilities of these disruptive technologies. Little is known about the robotics and computer vision industries in Australia today and yet they will support the use of robotics across a range of sectors including; manufacturing, services, healthcare, resources, the natural and built environments, and defence.

The Australian Centre for Robotic Vision is developing Australia’s first robotics and computer vision roadmap. The Centre has received in-depth advice on how to construct a nationally recognised roadmap fro Prof. Henrik Christensen, the author of the European and US robotics roadmaps. The 2013 version of the US robotics roadmap gained the attention of congress and President Barack Obama and, eventually, led to concentrated investment in robotics research in the US. Our goals for Australia are similar but also include computer vision and will tackle
the question of defining Australia’s capability in the space as well as future research directions.

We have an ambitious timeline to complete the roadmap for it to be launched in May 2018 in Brisbane, Australia, during the International Conference for Robotics and Automation (ICRA2018), the first time the
conference has been held in the southern hemisphere and a time when the spotlight will be on Australia’s robotics and computer vision communities. We are keen to represent diverse views in the roadmap and invite people to attend the workshop to contribute their ideas.