Laboratory for Intelligent and Flexible Machine
Mechanical Engineering Discipline, School of Engineering, Monash University
We are still far from creating intelligent machines, i.e. robots, with the same flexibility and adaptability as their biological counterparts. The primary goal of the research conducted in LIFE is the understanding of the principle found in biological systems that shows the importance of not only their computational intelligence, but also the embodied intelligence due to the flexible and soft body. Through the understanding, we envision the development of intelligent robots beyond the current technological advances in terms of flexibility, adaptability, and other related aspects like safe interaction and energy efficiency.
Surya Nurzaman (Monash University), Liyu Wang (Univ. of California Berkeley), Fumiya Iida (Univ. of Cambridge), Jeffrey Lipton (Univ. of Washington), Daniela Rus (MIT), Dario Floreano (EPFL)
Unlike most of today’s robots which are made of rigid materials, structures made of soft materials can be found everywhere in biological world, such as the muscles and skins. We investigate the use of soft and flexible materials in robotic systems, with the expectation to realise systems that are cheaper, safer and more adaptable than the level that the conventional rigid-material robots can achieve.
Through the inspiration from the design principles shown by biological systems, we envision the development of intelligent robotic systems that are closer to their biological counterparts in terms of flexibility, adaptability, safe interaction or energy efficiency. The principles are being investigated in mobile robotics, legged locomotion, aerial robotics, grasping, and wearable systems.
One of the primary bio-inspired principles being investigated in LIFE is embodied intelligence, sometimes also referred to as morphological computation. The principle essentially states that the size, shape and material properties of physically embodied systems, and their interaction with the environments, can facilitate control and sensing in generating intelligent behaviours.
One of the aspects where robots still cannot compete with their biological counterparts is energy efficiency. We investigate the underlying principles of energy efficiency in biological systems and applied them to different types of robots, in real life and simulation.
We also put significant effort into investigating and analysing the dynamics of robotic systems, particularly those at least partially made of soft and compliant materials. The effort includes self-organisation based approaches and model based approaches like mass-spring damper model and state estimation.
We are continuously engaging and seeking collaboration with industry to solve relevant technology problems, apply the investigated fundamental principles and promote the utilisation of newly discovered inventions and techniques.