Intelligent Robotics Program

Aiming to provide robot solutions for diverse fields, including manufacturing, agriculture, health care and civil engineering.

Led by Dr Mats Isaksson, the Intelligent Robotics program leverages Swinburne’s technology expertise in robotics, mechatronics, automatic control, computer vision, artificial intelligence and industry 4.0 to develop novel robotics technologies.

In close collaboration with our industrial partners we are developing purpose-built robots and novel automation solutions for a wide range of applications.

Research streams:

Developing purpose-built robots for unstructured environments, targeting applications in mining, farming, construction etc.

Targeting autonomous mobility platforms and artificial intelligence.

Improving the quality of human-robot interaction.

Researching novel robot solution and devices for health care applications.

 

Field Robotics

 

The aim of this stream is to develop purpose-built robots for unstructured environments, targeting applications in mining, farming, construction etc. The requirements for these applications typically involve both advanced robotics and mechatronics in addition to novel solutions for autonomy and control.

Researchers:

Projects: 

5-G Enabled Mobile Robots

This research aims to develop a network of mobile robots controlled over a 5G link in near real time. The project studied typical characteristics of 5G e.g., extremely low latency, high reliability and fast throughput and investigated the industrial applications to real-time control and automation enabled by 5G. The 5G-enabled mobile robots are demonstrated to accomplish tasks such as parts delivery, product inspection and collaboration with human operators or robots during the manufacturing process.

Autonomous Underwater Vehicles

Autonomous underwater vehicles (AUV) have potential applications in marine biology projects, dive site surveys, boat inspections, and underwater exploration. In the past few years, we have developed prototypes of AUV with the form of propeller-type and fish-like robot. Their functionality and performance are still being improved. The major goal of this project is to develop a real-time localisation system and autonomous navigation controller such that these underwater robots could follow a specified path. The research and development in progress is aimed for seafloor mapping and sea wave-induced vibration data collection with hydrophone.

Autonomous Agricultural Robots

This research aims to develop novel mobile robotic systems to operate in unstructured environment with specific focus on orchard, agricultural and forestry environments. Such a robotic system with on-board sensors including IMU and computer vision, superior mobility, agility and intelligence can automated a variety of operations such as assisting orchard workers, transporting, data collection, spraying, weeding, harvest, and environmental management. The research covers the following topics:

  • Novel locomotion and mobility of ground vehicle systems
  • Modelling and disturbance-free proximity control of Miniature Aerial Vehicles (MAV) for environment / infrastructure monitoring and management
  • Deep learning pattern recognition for robotic guidance in natural environments
  • Error modelling and quantification of imaging systems to improve localization
  • Multi-sensor data fusion for accurate localization and positioning
  • Multi-agent cooperative control for agricultural and post-farm gate logistics 

 

Artificial Intelligence and Autonomy

 

Within this research stream we target autonomous mobility platforms, including field sensing and perception. We are researching positioning and navigation solutions with a strong focus on machine learning, artificial intelligence, computer vision, multi-sensor data fusion, and autonomous navigation and control in unstructured and unknown environments. We are also researching multi-agent control and swarm intelligence.

Researchers:

Projects:

Real-time Video-based Passenger Analytics

Interest in Automatic Passenger Counting (APC) systems is rapidly growing as public transport providers increasingly seek accurate, real-time estimates of occupancy on the services they provide. Vision-based systems in particular represent a compelling low-cost and versatile choice.  However, various camera-based solutions exist representing design choices and trade-offs that must be understood in the context of specific public transport scenarios. Accuracy benchmarks using pre-existing data sets offer some guidance, however, must be supported by extensive in-field trialling if fitness for purpose is to be assessed. In this research, in partnership with Transport for New South Wales and iMOVE CRC, we are developing and evaluating real-time vision-based solutions for Automatic Passenger Counting. Leveraging state-of-the-art deep neural networks and visual tracking techniques, we have been exploring RGB image-based solutions, representing systems that may be deployable with existing CCTV cameras, as well as more novel sensor choices such as RGB+Depth solutions. Our research to date has evaluated our methods in both off-line testing, as well as in-field trials, including during peak-passenger metropolitan bus services in Sydney, Australia. Our solutions are also focussed on the use of low cost, low powered and easy to install solutions suitable for deployment on real passenger-facing public transport services.

  • Industry Partner: Transport for New South Wales, iMOVE CRC
  • Contact: Dr Chris McCarthy

Bio-inspired Vision Processing for Prosthetic Vision and Wearable Assistive Technologies

Retinal implants, also known as bionic eyes, are the leading treatment for retinal diseases such as Retinitis Pigmentosa and Age-related Macular Degeneration. This implantable micro-electrode array is able to restore a sense of vision by electrically stimulating surviving retinal cells. The primary input to this process are video frames acquired from a head-mounted camera. However, due to the highly restricted resolution of state-of-the-art implants (as low as 40 “pixels), a significant down-sampling must be performed on the originally captured image frames.  This motivates consideration of computer vision and image enhancement algorithms to detect, extract and augment the signal passed to the implant so as to emphasise key features in the environment that are crucial to safe and independent visual guidance for the implant recipient. In this research, we have explored the adaptation of robotic vision techniques such as ground plane segmentation, human vision inspired vision algorithms for motion processing, as well as state-of-the-art deep learning (and in particular deep reinforcement learning) to establish a data-driven pipeline for both learning and enhancing visual features of importance to particular high priority visuo-motor tasks such as navigation, and table-top object grasping. Through both patient testing, and simulated testing with virtual reality, we aim to develop and thoroughly evaluate vision processing solutions that aim to improve functional outcomes, enhance the independence and provide useful vision for patients implanted with such devices.

 

Human-Robot Interaction

 

This research stream aims to improve the quality of human-robot interaction. The research topics include collaborative robots (cobots), tele-robotics, task-oriented robot control, adapting to human, etc. Human-robot interaction is of growing importance with collaborative robots being used in more and more applications such as medical surgery, rehabilitation, entertainment and service.

Researchers:

Projects:

Tradiebot – Automatic Repair of Plastic Components in the Automotive Industry

Together with industrial partner Tradiebot Industries and supported by the Innovative Manufacturing Cooperative Research Centre (IMCRC), we are developing a revolutionising solution for automatic repair of headlight housings. The developed solution is an excellent example of an Industry 4.0 solution, integrating 3D scanning, 3D printing and advanced robotics with the development of a novel material. Several patent applications have already been submitted and an application for a follow up project is currently under review. 

Robot-Assisted Remote Echocardiographic Examination

Medical advances have continued to extend the life expectancy of the Australian population. One of the major contributions to this fact is early disease detection and rapid intervention. However, the access to health centres, hospitals and medical specialists enabling such early detection and intervention is very unequal depending on where in Australia a person lives. While people in major cities or urban areas typically have excellent access to the relevant facilities and specialists, people living in remote rural areas are significantly less fortunate. Sometimes the limitation is due to the cost of the required medical equipment, however, in many cases, the limitation is instead due to the low availability of medical specialists in remote areas.

One potential solution to this problem is the teleoperation of medical procedures. This technology is particularly important for a large, sparsely populated country like Australia. The proposed project is focused on utilising teleoperation for medical diagnosis and more specifically for remote heart ultrasounds. A heart echocardiography uses high-frequency sound waves to capture images of a heart enabling the cardiographer to study the heart’s functions such as the pumping strength, the working of the valves and to check if there are any tumours or infectious growth around the heart valves. An echocardiogram is a very important tool to diagnose heart conditions such as heart murmurs, enlarged heart, valve problems, etc.

This research project targets all aspects of robot-assisted remote ultrasound examinations.

Robot Control in Networked Environments

This project aims to provide a breakthrough in the understanding of dynamic behaviours of robotic control systems in networked environments, and develop a new theory for the analysis and design of networked robotic control subject to network induced constrains. The robotic control is used in many industry applications to improve efficiency, productivity, accuracy, etc. The development of digital technology, especially Industry 4.0, brings in more and more networked control systems, which creates new design challenges due to network induced constraints such as delay, data packet dropouts and cyber-attacks. This project is a Discovery Project sponsored by Australian Research Council.

Smart Robotic and Mechatronic Systems

In this project, we investigate smart algorithms for robotic and mechatronic systems subjected to nonlinearities, disturbances, uncertainties, etc. The tasks includes modelling and pattern recognition, control and optimization, sensoring and data processing. The developed algorithms have been used in electric vehicles steer by wire, anti-lock brake systems, flexible robotic joints, industrial drives, unmanned autonomous vehicles, etc.

 

Assistive Robots

 

We are researching novel robot solution and devices for health care applications. With the introduction of low-cost speed and power limited collaborative robots, we believe intelligent robot solutions are set to revolutionize this area. Other focus areas within this stream include bionic robotic device, bio-sensing, brain machine interface and biomechanics modelling and control.

Researchers:

Projects:

A Collaborative Robot System for Photobiomodulation Therapy of Chronic Pain

‌In a partnership with IR Robotics Pty Ltd, we have developed a solution to automate photobiomodulation therapy of chronic pain. Integrating a thermal camera, a low level infrared laser and a novel collaborative robot from ABB Robotics, our solution can automatically and safely perform photobiomodulation therapy on patients with soft tissue injury. Building on Industry 4.0 technologies, the derived platform has the potential to self-adapt to provide individually optimal treatment.

Robot-Assisted Laparoscopic Surgery – An Ergonomic Investigation

‌Assisted Laparoscopic Surgery (RALS) is known to provide a range of benefits to the patient. In this project, we are instead focusing on the benefits to the surgeon, aiming to prove to which extent RALS can extend a surgeon’s career length. The research project is a collaboration between Swinburne University and Healthe Care, aiming to quantify the ergonomic benefits to the surgeon when using robot-assisted laparoscopic surgery compared to traditional laparoscopic surgery.  We are researching sensor fusion technologies, combining EMG measurements with motion capture data and advanced biomechanical modelling

Socially Assistive Robots for Paediatric Rehabilitation

Swinburne University of Technology, in partnership with Melbourne’s Royal Children's Hospital, has been developing a socially assistive humanoid robot as a therapeutic aid for paediatric rehabilitation, aiming to motivate young patients, increase exercise adherence, and deliver rehabilitation care efficiently. After 3.5 years of development and testing, the robot has engaged with over 50 patients, predominantly with cerebral palsy, and is undergoing evaluation with therapists and parents within RCH’s rehabilitation clinic. While much has been achieved, adapting and deploying a general purpose social robot as a therapeutic aid for diverse young patients, with varying needs, in a busy hospital setting, comes with significant challenges  not typically addressed in traditional Human-Robot Interaction or robotics research generally. In this research we are exploring how state-of-the-art artificial intelligence and computer vision can be deployed long term in such real-world, clinical settings, to augment therapy delivery and improve patient outcomes.

  • Industry Partner: Royal Children’s Hospital, Melbourne and The Brainary
  • Contact: Dr Chris McCarthy

Assist-as-Needed Robotic Rehabilitation Device

The rising cost of medical treatment and the rapidly aging world population has necessitated the advent of cheaper alternative methods for rehabilitation therapy and assistive technology. The advancement of robotic technology has brought about the possibility of robotic alternative to the above-mentioned concerns, and will potentially bridge the current treatment gap and provide better assistive orthosis. The project aims to design and develop an intelligent rehabilitative/assistive orthotic device for lower limbs and upper limbs, intended to aid stroke victims and also those with limited limb mobility due to weak muscles. The research covers the following topics:

  • Design, modelling and control of artificial muscle and smart actuator
  • Assist as needed adaptive control to support and reinforce natural movements of limbs through the range of motion (ROM)
  • Intent detection and feedback assistance control based on bio-sensing
  • Bio-sensing and instrumentation for patient condition monitoring and diagnosis

Designing Robot Interactions for Engagement with and for Older Adults

As a team of interaction designers, psychologists and engineers, we are interested in the relationship with and attitudes towards robots. We explore which characteristics increase acceptance and adoption in everyday life creating perceived benefits for the user.  Also ethical considerations and societal issues surrounding the role of robots are researched and discussed - in particular in the topic of social robotics. We work with active ageing groups from the City of Wyndham in the domains of health and wellbeing through robots and what they can contribute to quality of life for older adults and adults living with dementia.

Feeling at Home with Your Robot

Traditionally a strong focus in robot development concerns the functional aspects of robots. The design aesthetics of robots and their outputs (speech, gestures and emotions) are often overlooked or neglected. We are interested in speech interactions, the visual appearance of robots and their ability to convey emotions. We explore the impact of improved speech interfaces and communication patterns of robots with users and how these support older adults to live independently at home for longer. 

Contact the Manufacturing Futures Research Institute

There are many ways to engage with us. If your organisation is dealing with a complex problem, then get in touch to discuss how we can work together to provide solutions.

Call +61 3 9214 5177