Artificial intelligence and robotics on CSE Robotics

Blinky and the 2024 software system

Mon, 01 Jan 0001 00:00:00 +0000

The Toyota HSR platform

rUNSWeep’s earlier RoboCup@Home work used the Toyota HSR-B, known as Blinky. The team’s original account describes a domestic-robotics testbed in the Human Robot Interaction Lab, set up as a small apartment. The team was established in 2016 and first competed at RoboCup@Home in Nagoya in 2017.

How the 2024 software fitted together

The original system description grouped its ROS packages by capability:

Claude Sammut

Mon, 01 Jan 0001 00:00:00 +0000

Claude Sammut’s research connects machine learning with robotics. His work includes relational learning and behavioural cloning, alongside the development of autonomous robot behaviour. He has led UNSW teams across several generations of RoboCup.

RoboCup Past Teams

Mon, 01 Jan 0001 00:00:00 +0000

2023 Bordeaux, France

Standard Platform League “rUNSWift”

World: 3rd Place

Domestic Standard Platform League “UNSW@Home”

World:

2022 Bangkok, Thailand

Standard Platform League “rUNSWift”

World: 3rd Place

Domestic Standard Platform League “UNSW@Home”

World:

2019 Sydney,Australia

Standard Platform League “rUNSWift”

World: 3rd Place

Domestic Standard Platform League “UNSW@Home”

World:

2018 Montreal,Canada

Robot learning and autonomous behaviour

Mon, 01 Jan 0001 00:00:00 +0000

A robot needs to adapt when its environment, task or physical behaviour changes. Learning offers ways to acquire useful behaviours from data and experience, while control connects those behaviours to the constraints of a physical system.

CSE work has combined learning from demonstration, explanation-based learning and reinforcement learning. Observations can supply an abstract account of a behaviour, which then guides trial-and-error learning. Earlier systems applied this approach to bipedal locomotion, rescue robots and learning to use objects as tools—for example, using a hook to retrieve an object beyond the robot’s reach.

rUNSWift

Mon, 01 Jan 0001 00:00:00 +0000

Autonomous robot soccer

rUNSWift brings together students and researchers to build robots that play soccer autonomously. The task connects vision, localisation, motion, decision-making and coordination in a single physical system.

Building a complete autonomous system

The original team material identifies vision, motion, localisation, behaviours, communication, system architecture, tooling and robot maintenance as connected areas of work. On a soccer field, progress in one component only helps when it works with the rest of the system.

Oliver Obst

Mon, 01 Jan 0001 00:00:00 +0000

Oliver Obst works on embodied intelligence and collective robotics. His research includes efficient learning systems, recurrent neural networks, multi-robot coordination and autonomous platforms operating under uncertainty.

Reasoning and cognitive robotics

Mon, 01 Jan 0001 00:00:00 +0000

Autonomous robots need to connect what they observe with what they know, what they want to achieve and what they can do next. Cognitive robotics studies the representations and reasoning processes that support these decisions.

Work in this theme includes reasoning about actions, task planning, knowledge representation and high-level robot programming. CSE’s earlier blackboard architectures used shared memory to connect cognitive processes. Cognitive hierarchies provided a framework for integrating symbolic and sub-symbolic representations, while action formalisms supported planning and reasoning about incomplete knowledge.

rUNSWeep

Mon, 01 Jan 0001 00:00:00 +0000

Robots in everyday environments

rUNSWeep’s RoboCup@Home work brings together navigation, manipulation, dialogue and task reasoning in a domestic setting. These tasks require a robot to connect observations with a model of objects, places and people.

An integrated domestic-robotics testbed

The original rUNSWeep material describes ROS software connecting perception, a world model, navigation, manipulation, speech and task execution. Tasks such as cleaning a table and carrying luggage make the integration problem concrete: observations must lead to a useful sequence of actions, with feedback as the situation changes.

Francisco Cruz

Mon, 01 Jan 0001 00:00:00 +0000

Francisco Cruz works on reinforcement learning, explainable artificial intelligence and human–robot interaction. His research includes interactive approaches in which people can help shape the behaviour of learning systems.

Perception and human alignment

Mon, 01 Jan 0001 00:00:00 +0000

Robots act on incomplete observations. Computer vision and multimodal perception help a robot recognise objects, interpret scenes and relate its actions to the people around it. Human–robot interaction asks how those capabilities can support understandable and useful behaviour.

This theme connects visual understanding, interactive learning and human-centred autonomy. It includes questions about how a robot can interpret feedback, communicate its intentions and adapt its behaviour to a shared task. The physical setting matters: working near a person places demands on perception, movement and the predictability of the whole system.

RoboCup Qualification

Mon, 01 Jan 0001 00:00:00 +0000

2026 Incheon, South Korea

Team Description Paper rUNSWeep_TDP_2026

Watch the qualification video on YouTube

2025 Salvador, Brazil

Team Description Paper rUNSWeep_TDP_2025

Watch the qualification video on YouTube

2024 Eindhoven, Netherlands

Team Description Paper UNSW DSPL TDP 2024

Poster

Watch the qualification video on YouTube

2023 Bodeaux, France

Team Description Paper rUNSWeep_TDP_2023

Poster rUNSWeep_poster_2023

Watch the qualification video on YouTube

2022 Bangkok, Thailand

Team Description Paper

Poster rUNSWeep_poster_2022

2019 Sydney, Australia

Team Description Paper rUNSWeep_TDP_2019

RoboCup Rescue

Mon, 01 Jan 0001 00:00:00 +0000

Robots for search and rescue

RoboCup Rescue develops autonomous search-and-rescue capabilities for simulated disaster environments. The team works with a Unitree Go2 quadruped, bringing together mapping, navigation and computer vision to explore unfamiliar terrain and identify potential victims.

Connecting perception and action

The project uses ROS 2 to integrate sensors, localisation, planning and robot control. It forms part of UNSW’s AI 4 Everyone Vertically Integrated Project, with Yang Song and Maurice Pagnucco as academic leads.

rUNSWift – History

Mon, 01 Jan 0001 00:00:00 +0000

Competition Results
Standard Platform League
4-Legged League
Robots Timeline

With the competition being “Open Source” (the source code is published at the end of the year so that AI improves all around the world), UNSW has to work hard at not only gain an edge, but to also out-do themselves each year. The core team of UNSW RoboCup uses a mix of university students and collaborators, all of whom have to learn about Robocup’s technology from scratch before contributing to the team. Our team members have produced a number of publications which have been crucial in our success.

Michael Thielscher

Mon, 01 Jan 0001 00:00:00 +0000

Michael Thielscher’s research includes cognitive robotics, knowledge representation, intelligent agents and general game playing. His work addresses how an autonomous system represents knowledge and reasons about its actions.

Multi-robot systems and embodied autonomy

Mon, 01 Jan 0001 00:00:00 +0000

A team of robots must make decisions while sharing a physical environment. Coordination depends on what each robot can observe, communicate and do, and on how those capabilities change during a task.

This theme connects multi-agent systems, collective robotics and embodied intelligence. Research questions concern how individual decisions produce useful team behaviour, how robots cope with uncertainty and how sensing and communication constrain coordination. Physical testbeds expose interactions that can be difficult to capture in a simplified model.

History

Mon, 01 Jan 0001 00:00:00 +0000

RoboCup @Home Domestic Standard Platform League (DSPL)

rUNSWeep compete in the Domestic Standard Platform League. The team was established in 2016 when we received our Toyota HSR-B. The team first competed in the 2017 RoboCup@Home in Nagoya, Japan.


	RoboCup 2024 – Eindhoven, Netherlands 17-21 July 2024
	RoboCup 2023 – Bordeaux, France 4-10 July 2023
	RoboCup 2022 – Bangkok, Thailand 13-17 July 2022
	RoboCup 2019 – Sydney, Australia 2-8 July 2019
	RoboCup 2018 – Montreal, Canada 18-22 June 2018
	RoboCup 2017 – Nagoya, Japan 27-31 July 2017

Maurice Pagnucco

Mon, 01 Jan 0001 00:00:00 +0000

Maurice Pagnucco works in knowledge representation and reasoning, cognitive robotics, belief change and reasoning about actions. Follow his UNSW profile for publications and current institutional roles.

NAO robot soccer platform

Mon, 01 Jan 0001 00:00:00 +0000

An earlier generation of robot soccer

rUNSWift’s Standard Platform League work used NAO humanoid robots. Using a shared hardware platform focused competition on perception, localisation, motion and autonomous behaviour.

The team won the Standard Platform League in 2014 and 2015. In 2026, rUNSWift competed with Booster K1 robots. This entry records the earlier platform and does not describe the current team’s hardware.

See UNSW’s historical competition results and the team archive.

Yang Song

Mon, 01 Jan 0001 00:00:00 +0000

Yang Song’s research spans computer vision, deep learning, multimodal and trustworthy AI. Her UNSW profile provides current research outputs, supervision information and contact details.

Germán Castro

Mon, 01 Jan 0001 00:00:00 +0000

Dr Germán Castro’s research areas include intelligent robotics, artificial intelligence, machine learning and autonomous vehicle systems. His UNSW Research profile is listed under the full name German Javier Castro Donoso and provides links to his research outputs and ORCID record.

Projects

rUNSWeep: service robotics and RoboCup@Home

UNSW@Work

Mon, 01 Jan 0001 00:00:00 +0000

The 2024 project record

The original GitLab site described UNSW@Work’s aim of developing a holonomic mobile robot for RoboCup@Work 2024. The proposed system would navigate a simulated factory floor, collect objects and deliver them using a robotic arm.

The project brought together LiDAR and depth sensing, SLAM-based navigation and computer vision for recognising objects. Its research setting connected perception and mapping with physical manipulation in an industrial-style task.

Mahdi Bamdad

Mon, 01 Jan 0001 00:00:00 +0000

Mahdi Bamdad is a Senior Technical Officer in UNSW’s School of Computer Science and Engineering. His research publications include work on cable-driven robots, robot mechanisms and control, and rehabilitation robotics.

His ORCID record provides links to his research outputs.

Software Architectures for Autonomous Robots

Mon, 01 Jan 0001 00:00:00 +0000

An example cognitive hierarchy

The CSE robotics group was an early proponent of blackboard architectures for sharing data between a robot’s cognitive processes. A blackboard provides a global shared memory facility that acts as both a communications infrastructure as well as a storage mechanism for a robot. The blackboard typically offers both publisher-subscriber based asynchronous communications as well as synchronous query-answering communications. The storage of the blackboard provide both short-term and long-term memory.

Learning Robot Behaviours

Mon, 01 Jan 0001 00:00:00 +0000

Watch the qualification video on YouTube

Demonstration of a robot learning the correct tool to use to solve a particular task.

Watch the qualification video on YouTube

Demonstration of a robot learning to climb a low-step, a high-step and staircase.

The work documented here developed machine learning techniques for robotics. Manually programming new robot behaviours is very challenging. A robot that can learn its own behaviours can better and more quickly adapt to new situations and tasks, without relying on domain experts and programmers. This learning is very beneficial for robots for home environments, as end users could instruct their robots to learn new skills, rather than rely and wait on teams of experts to design and implement new software.

Robot Control and Reasoning

Mon, 01 Jan 0001 00:00:00 +0000

Watch the qualification video on YouTube

Demonstration of a robot learning to climb a low-step, a high-step and staircase.

Our group has a strong background in formal logic based AI and particularly in the area of reasoning about knowledge and action. The expertise in this area extends from research into abstract representation formalisms for actions and knowledge, through to pragmatic issues of implementing reasoning for the purposes of controlling robot behaviour.

Position Tracking and SLAM

Mon, 01 Jan 0001 00:00:00 +0000

2D SLAM Map of the CSE Robotics Lab

Our research group has a long history of research in position tracking and SLAM, particularly for autonomous robots in the RoboCupRescue Robot competition. The work described here focused on implementing position tracking and SLAM algorithms on GPUs using full 3D information to produce correctly aligned and accurate 3D maps.

Selected publications from this work

A. Ratter, C. Sammut. Fused 2D/3D Position Tracking for Robust SLAM on Mobile Robots. International Conference on Intelligent Robots and Systems, 2015.
A. Ratter, C. Sammut. Local Map Based Graph SLAM with Hierarchical Loop Closure and Optimisation. Australasian Conference on Robotics and Automation, 2015.
A. Ratter and C. Sammut, GPU Accelerated Parallel Occupancy Voxel Based ICP for Position Tracking, 2013 Australasian Conference on Robotics and Automation, Sydney, Australia, pp. 394–403, 2013.
A. Ratter, C. Sammut, M. McGill. GPU Accelerated Graph SLAM and Occupancy Voxel Based ICP For Encoder-Free Mobile Robots. International Conference on Intelligent Robots and Systems, 2013
A. Milstein, M. McGill, T. Wiley, R. Salleh, and C. Sammut, A Method for Fast Encoder-Free Mapping in Unstructured Environments, Journal of Fields Robotics: Special Issue on Safety, Security, and Rescue Robotics, vol. 28, no. 6, pp. 817–831, 2011
A. Milstein, M. McGill, T. Wiley, R. Salleh, and C. Sammut, Occupancy voxel metric based iterative closest point for position tracking in 3D environments, IEEE International Conference on Robotics and Automation, pp. 4048 – 4053, 2011.

3D SLAM Map of the CSE Robotics Lab

General Game Playing Robots

Mon, 01 Jan 0001 00:00:00 +0000

Our research group has particular expertise in the area of General Game-Playing (GGP) and we have been active in extending this notion to the robotics domain. GGP is the attempt to create AI systems that can understand and learn to play new games without human intervention. Extending GGP to the robotics domain requires moving beyond a virtual/abstract environment to consideration of game playing within an embodied physical environment.

For example, a domestic robot tasked with fetching an item has to consider the possibility that the item may not be where it expects, or that the human operator may change locations after issuing the request. Viewing such a task as a game can provide a framework for developing more natural and intuitive robot behaviours.

Autonomous Adaptation and Trust

Mon, 01 Jan 0001 00:00:00 +0000

Trust becomes a critical issue for humans working with robots, especially when they can autonomously learn and adapt to new situations. The behaviour of these types of machines cannot be formally verified in advance. The original proposal set out to study the change in trust for a mixed initiative task under varying degrees of transparency of the adaptation process.

The proposal identified two intended contributions:

The design and development of a robotic cognitive architecture that includes the ability for the robot to adapt autonomously to a change in the task environment. The proposed architecture used a Baxter robot for participation in a mixed initiative task where the environment changes, requiring the robot to adapt on the job.
Modelling and evaluating the evolving human-robot trust relationship as the robot learns on the job.

rUNSWift tests a new generation of humanoid robots

Tue, 07 Jul 2026 00:00:00 +0000

At RoboCup 2026, rUNSWift competed with Booster K1 humanoid robots for the first time. Preparing the new platform involved adapting locomotion, localisation, game play and ROS 2 software.

The competition exposed challenges in integrating those components under match conditions and gave the team a clearer basis for its next stage of work.

Read the report from the Embodied Intelligence & Collective Robotics Lab · Explore rUNSWift

HSR has arrived

Sun, 02 Apr 2017 11:32:35 +1000

Our HSR (on loan from Toyota) has arrived!