UTS site search

Dr Benjamin Johnston

Biography

For current information about Benjamin Johnston including biography, publications and research interests, please refer to his personal webpage.

Benjamin is best reached by email. Please do not call the phone number listed.

Senior Lecturer, School of Software
Core Member, QCIS - Quantum Computation and Intelligent Systems
Core Member, AAI - Advanced Analytics Institute
PhD (UTS)
 
Phone
+61 2 9514 1851

Conferences

Williams, M., Raza, R.A. & Johnston, B. 2016, 'Reward from Demonstration in Interactive Reinforcement Learning', Proceedings of the Twenty-Ninth International Florida Artificial Intelligence Research Society Conference.
Vitale, J., Williams, M.-.A. & Johnston, B. 2016, 'The face-space duality hypothesis: a computational model', Proceedings of the 38th Annual Conference of the Cognitive Science Society, 38th Annual Meeting of the Cognitive Science Society, Cognitive Science Society, Philadelphia, pp. 514-519.
View/Download from: UTS OPUS
Romat, H., Williams, M.-.A., Wang, X., Johnston, B., Bard, H. & IEEE 2016, 'Natural Human-Robot Interaction Using Social Cues', Proceedings of the HRI '16 The Eleventh ACM/IEEE International Conference on Human Robot Interaction, ACM/IEEE International Conference on Human Robot Interaction, ACM, Christchurch, New Zealand, pp. 503-504.
View/Download from: UTS OPUS
This paper investigates the problem of how humans understand and control human-robot collaborative action and how to build natural interactions during human-robot collaborative action. We use a "pick and place" experiment to study collaborative activities between a human and a robot. The results show that even if human participants had a good understanding of the maximum reachability of the robot, they consistently take a surprisingly long time to help and assist the robot when a target object is out of its reach. We implemented a number of social cues in the experiment, analysed their effects in order to identify the role they could play to improve the fluency of human-robot collaboration. The experimental results showed that when the robot uses head movements, two hands or a gesture to indicate non-reachability, people react in a more natural way to assist the robot.
Leong, T.W. & Johnston, B. 2016, 'Co-design and Robots: A Case Study of a Robot Dog for Aging People', SOCIAL ROBOTICS, (ICSR 2016), pp. 702-711.
View/Download from: UTS OPUS or Publisher's site
Romat, H., Williams, M.-.A., Wang, X., Johnston, B., Bard, H. & ACM 2016, 'Natural Human-Robot Interaction Using Social Cues', Proceedings of the 11th ACM/IEEE International Conference on Human Robot Interaction, ACM/IEEE International Conference on Human Robot Interaction (HRI), IEEE, Christchurch, New Zealand, pp. 503-504.
This paper investigates the problem of how humans understand and control human-robot collaborative action and how to build natural interactions during human-robot collaborative action. We use a "pick and place" experiment to study collaborative activities between a human and a robot. The results show that even if human participants had a good understanding of the maximum reachability of the robot, they consistently take a surprisingly long time to help and assist the robot when a target object is out of its reach. We implemented a number of social cues in the experiment, analysed their effects in order to identify the role they could play to improve the fluency of human-robot collaboration. The experimental results showed that when the robot uses head movements, two hands or a gesture to indicate non-reachability, people react in a more natural way to assist the robot.
Vitale, J., Williams, M.-.A. & Johnston, B. 2014, 'Socially impaired robots: Human social disorders and robots' socio-emotional intelligence', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 6th International Conference on Social Robotics, Springer Verlag, Sydney, Australia, pp. 350-359.
View/Download from: UTS OPUS or Publisher's site
Social robots need intelligence in order to safely coexist and interact with humans. Robots without functional abilities in understanding others and unable to empathise might be a societal risk and they may lead to a society of socially impaired robots. In this work we provide a survey of three relevant human social disorders, namely autism, psychopathy and schizophrenia, as a means to gain a better understanding of social robots' future capability requirements.We provide evidence supporting the idea that social robots will require a combination of emotional intelligence and social intelligence, namely socio-emotional intelligence. We argue that a robot with a simple socio-emotional process requires a simulation-driven model of intelligence. Finally, we provide some critical guidelines for designing future socio-emotional robots.
Wang, X., Williams, M.-.A., Gardenfors, P., Vitale, J., Abidi, S., Johnston, B., Kuipers, B. & Huang, A. 2014, 'Directing human attention with pointing', Robot and Human Interactive Communication, 2014 RO-MAN: The 23rd IEEE International Symposium on, The 23rd IEEE International Symposium on Robot and Human Interactive Communication, IEEE, Edinburgh, Scotland, pp. 174-179.
View/Download from: UTS OPUS or Publisher's site
Pointing is a typical means of directing a human's attention to a specific object or event. Robot pointing behaviours that direct the attention of humans are critical for human-robot interaction, communication and collaboration. In this paper, we describe an experiment undertaken to investigate human comprehension of a humanoid robot's pointing behaviour. We programmed a NAO robot to point to markers on a large screen and asked untrained human subjects to identify the target of the robots pointing gesture. We found that humans are able to identify robot pointing gestures. Human subjects achieved higher levels of comprehension when the robot pointed at objects closer to the gesturing arm and when they stood behind the robot. In addition, we found that subjects performance improved with each assessment task. These new results can be used to guide the design of effective robot pointing behaviours that enable more effective robot to human communication and improve human-robot collaborative performance.
Novianto, R., Williams, M.-.A., Gärdenfors, P. & Wightwick, G. 2014, 'Classical conditioning in social robots', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Springer Verlag, Germany, pp. 279-289.
View/Download from: UTS OPUS or Publisher's site
Classical conditioning is important in humans to learn and predict events in terms of associations between stimuli and to produce responses based on these associations. Social robots that have a classical conditioning skill like humans will have an advantage to interact with people more naturally, socially and effectively. In this paper, we present a novel classical conditioning mechanism and describe its implementation in ASMO cognitive architecture. The capability of this mechanism is demonstrated in the Smokey robot companion experiment. Results show that Smokey can associate stimuli and predict events in its surroundings. ASMO's classical conditioning mechanism can be used in social robots to adapt to the environment and to improve the robots' performances.
Abidi, S.S., Williams, M. & Johnston, B.G. 2013, 'Human pointing as a robot directive', ACM/IEEE International Conference on Human-Robot Interaction, ACM/IEEE International Conference on Human-Robot Interaction, IEEE, Tokyo, Japan, pp. 67-68.
View/Download from: Publisher's site
People are accustomed to directing other people's attention using pointing gestures. People enact and interpret pointing commands often and effortlessly. If robots understand human intentions (e.g. as encoded in pointing-gestures), they can reach higher
Felix Navarro, K.M., Gay, V.C., Johnston, B.G., Leijdekkers, P., Vaughan, E.P., Wang, T. & Williams, M. 2013, 'SocialCycle What Can a Mobile App Do To Encourage Cycling', 38th IEEE Conference on Local Computer Networks (LCN 2013) and Workshops, Second IEEE International Workshop on Global Trends in Smart Cities 2013, IEEE Computer Society, Sydney Australia, pp. 24-30.
View/Download from: UTS OPUS
Traffic congestion presents significant enviromnental, social and economic costs. Encouraging people to cycle and use other fonns of alternate transportation is one important aspect of addressing these problems. However, many city councils face significant difficulties in educating citizens and encouraging them to fonn new habits around these alternate fonns of transport. Mobile devices present a great opportunity to effect such positive behavior change. In this paper we discuss the results of a survey aimed at understanding how mobile devices can be used to encourage cycling and/or improve the cycling experience. We use the results of the survey to design and develop a mobile app called SocialCycle, which purpose is to encourage users to start cycling and to increase the number of trips that existing riders take by bicycle
Williams, M.-.A., Abidi, S., Gaerdenfors, P., Wang, X., Kuipers, B. & Johnston, B. 2013, 'Interpreting Robot Pointing Behavior', SOCIAL ROBOTICS, ICSR 2013, pp. 148-159.
View/Download from: UTS OPUS
Wang, W., Johnston, B.G. & Williams, M. 2012, 'Social networking for robots to share knowledge, skills and know-how', International Conference on Social Robotics, Springer, Chengdu, China, pp. 418-427.
View/Download from: UTS OPUS
A major bottleneck in robotics research and development is the difficulty and time required to develop and implement new skills for robots to realize task-independence. In spite of work done in terms of task model transfer among robots, so far little work has been done on how to make robots task-independent. In this paper, we describe our work-in-progress towards the development of a robot social network called Numbots that draws on the principle of sharing information in human social networking. We demonstrate how Numbots has the potential to assist knowledge sharing, know-how and skill transfer among robots to realize task-independence.
Johnston, B.G. 2011, 'The Collection of Physical Knowledge and Its Application in Intelligent Systems', Artificial General Intelligence - Lecture Notes in Compter Science: Proceedings of the 4th International Conference, AGI 2011, Artificial General Intelligence, Springer, Mountain View, USA, pp. 163-173.
View/Download from: UTS OPUS or Publisher's site
Intelligence is a multidimensional problem of which physical reasoning and physical knowledge are important dimensions. However, there are few resources of physical knowledge that can be used in data-driven approaches to Artificial Intelligence. Comirit Objects is a project intended to encourage the general public to contribute to research in Artificial Intelligence by building simple 3D models of everyday objects via an interactive web-site. This paper describes the simplified representation and web-interface used by Comirit Objects and a preliminary investigation into the potential applications of the collected models
Johnston, B.G. 2011, 'An Interface for Crowd-sourcing Spatial Models of Commonsense', Commonsense 2011, Symposium on Logical Formalizations of Commonsense Reasoning, AAAI Press, Stanford University, pp. 1-4.
View/Download from: UTS OPUS
Commonsense is a challenge not only for representation and reasoning but also for large scale knowledge engineering required to capture the breadth of our `everyday world. One approach to knowledge engineering is to `outsource the effort to the public through games that generate structured commonsense knowledge from user play. To date, such games have focused on symbolic and textual knowledge. However, an effective commonsense reasoning system will require spatial and physical reasoning capabilities. In this paper, I propose a tool for gathering commonsense information from ordinary people. It is a user-friendly 3D sculpting tool for modeling and annotating models of physical objects and spaces.
Williams, M., Gardenfors, P., Johnston, B.G. & Wightwick, G.R. 2010, 'Anticipation as a Strategy: A Design Paradigm for Robotics', Lecture Notes in Artificial Intelligence 6291 - Knowledge Science, Engineering and Management, Knowledge Science, Engineering and Management, Springer-Verlag Berlin Heidelberg, Belfast, Northern Ireland, pp. 341-353.
View/Download from: UTS OPUS or Publisher's site
Anticipation plays a crucial role during any action, particularly in agents operating in open, complex and dynamic environments. In this paper we consider the role of anticipation as a strategy from a design perspective. Anticipation is a crucial skill in sporting games like soccer, tennis and cricket. We explore the role of anticipation in robot soccer matches in the context of reaching the RoboCup vision to develop a robot soccer team capable of defeating the FIFA World Champions in 2050. Anticipation in soccer can be planned or emergent but whether planned or emergent, anticipation can be designed. Two key obstacles stand in the way of developing more anticipatory robot systems; an impoverished understanding of the âanticipationâ process/capability and a lack of know-how in the design of anticipatory systems. Several teams at RoboCup have developed remarkable preemptive behaviors. The CMU Dive and UTS Dodge are two compelling examples. In this paper we take steps towards designing robots that can adopt anticipatory behaviors by proposing an innovative model of anticipation as a strategy that specifies the key characteristics of anticipation behaviors to be developed. The model can drive the design of autonomous systems by providing a means to explore and to represent anticipation requirements. Our approach is to analyze anticipation as a strategy and then to use the insights obtained to design a reference model that can be used to specify a set of anticipatory requirements for guiding an autonomous robot soccer system.
Novianto, R., Johnston, B.G. & Williams, M. 2010, 'Attention in the ASMO cognitive architecture', Biologically Inspired Cognitive Architectures 2010 - Frontiers in Artificial Intelligence and Applications vol 221: Proceedings of the First Annual Meeting of the BICA Society, Annual Meeting of the BICA Society, IOS Press, Washington, USA, pp. 98-105.
View/Download from: UTS OPUS or Publisher's site
The ASMO Cognitive Architecture has been developed to support key capabilities: attention, awareness and self-modification. In this paper we describe the underlying attention model in ASMO. The ASMO Cognitive Architecture is inspired by a biological attention theory, and offers a mechanism for directing and creating behaviours, beliefs, anticipation, discovery, expectations and changes in a complex system. Thus, our attention based architecture provides an elegant solution to the problem of behaviour development and behaviour selection particularly when the behaviours are mutually incompatible.
Johnston, B.G. 2010, 'The toy box problem (and a preliminary solution)', Artificial General Intelligence - Proceedings of the Third Conference on Artificial General Intelligence, AGI 2010, Conference on Artificial General Intelligence, Atlantis Press, Lugano, Switzerland, pp. 43-48.
View/Download from: UTS OPUS or Publisher's site
The evaluation of incremental progress towards 'Strong AI' or 'AGI' remains a challenging open problem. In this paper, we draw inspiration from benchmarks used in artificial commonsense reasoning to propose a new benchmark problem- the Toy Box Problem-th
Johnston, B.G. & Williams, M. 2009, 'Conservative and Reward-driven Behavior Selection in a Commonsense Reasoning Framework', 2009 AAAI Symposium: Multirepresentational Architectures for Human-Level Intelligence, National Conference of the American Association for Artificial Intelligence, AAAI Press, Washington, USA, pp. 14-19.
View/Download from: UTS OPUS
Comirit is a framework for commonsense reasoning that combines simulation, logical deduction and passive machine learning. While a passive, observation-driven approach to learning is safe and highly conservative, it is limited to interaction only with those objects that it has previously observed. In this paper we describe a preliminary exploration of methods for extending Comirit to allow safe action selection in uncertain situations, and to allow reward-maximizing selection of behaviors.
Johnston, B.G. & Williams, M. 2009, 'A Formal Framework for the Symbol Grounding Problem', Proceedings of the Second Conference on Artificial General Intelligence, Conference on Artificial General Intelligence, Atlantis Press, Washington, USA, pp. 61-66.
View/Download from: UTS OPUS or Publisher's site
A great deal of contention can be found within the published literature on grounding and the symbol grounding problem, much of it motivated by appeals to intuition and unfalsifiable claims. We seek to define a formal framework of representa- tion grounding that is independent of any particular opinion, but that promotes classification and comparison. To this end, we identify a set of fundamental concepts and then formalize a hierarchy of six representational system classes that corre- spond to different perspectives on the representational require- ments for intelligence, describing a spectrum of systems built on representations that range from symbolic through iconic to distributed and unconstrained. This framework offers utility not only in enriching our understanding of symbol grounding and the literature, but also in exposing crucial assumptions to be explored by the research community.
Johnston, B.G. & Williams, M. 2009, 'Autonomous Learning of Commonsense Simulations', International Symposium on Logical Formalizations of Commonsense Reasoning, Symposium on Logical Formalizations of Commonsense Reasoning, UTSePress, Toronto, Canada, pp. 73-78.
View/Download from: UTS OPUS
Parameter-driven simulations are an effective and efficient method for reasoning about a wide range of commonsense scenarios that can complement the use of logical formalizations. The advantage of simulation is its simplified knowledge elicitation process: rather than building complex logical formulae, simulations are constructed by simply selecting numerical values and graphical structures. In this paper, we propose the application of machine learning techniques to allow an embodied autonomous agent to automatically construct appropriate simulations from its real-world experience. The automation of learning can dramatically reduce the cost of knowledge elicitation, and therefore result in models of commonsense with breadth and depth not possible with traditional engineering of logical formalizations.
Johnston, B.G. & Williams, M. 2008, 'Comirit: Commonsense Reasoning by Integrating Simulation and Logic', Artificial General Intelligance 2008 Proceedings of the First AGI Conference, Conference on Artificial General Intelligence, IOS Press, Inst of Technology, University of Memphis, TN, USA, pp. 200-211.
View/Download from: UTS OPUS
Rich computer simulations or quantitative models can enable an agent to realistically predict real-world behaviour with precision and performance that is difficult to emulate in logical formalisms. Unfortunately, such simulations lack the deductive flexibility of techniques such as formal logics and so do not find natural application in the deductive machinery of commonsense or general purpose reasing systems. This dilemma can, however, be resolved via a hybrid architecture that combines tableaux-based reasoning with a framework for generic simulation based on the concept of 'molecular' models. This combination exploits the complementary strengths of logic and simulation, allowing an agent to build and reason with automatically constructed simulations in a problem-sensitive manner.
Johnston, B.G., Yang, F., Mendoza, R., Chen, X. & Williams, M. 2008, 'Ontology Based Object Categorization for Robots', Lecture Notes in Artificial Intelligence Vol 5345: Practical Aspects of Knowledge Management - Proceedings of the 7th International Conference, PAKM 2008, International Conference on Practical Aspects of Knowledge Management, Springer, Yokohama, Japan, pp. 219-231.
View/Download from: UTS OPUS or Publisher's site
Meaningfully managing the relationship between representations and the entities they represent remains a challenge in robotics known as grounding. In this paper we Semantic Web technologies to provide a powerful extension to existing proposals for grounding robotic systems and have consequently developed OBOC, the first robotic software system with an ontology-based vision syb-system.
Mendoza, R., Johnston, B.G., Yang, F., Huang, Z., Chen, X. & Williams, M. 2007, 'OBOC: Ontology Based Object Categorisation for Robots', The Fourth International Conference on Computational Intelligence, Robotics and Autonomous Systems, International Conference on Computational Intelligence, Robotics and Autonomous Systems, Massey University Press, Palmerston North, New Zealand, pp. 178-183.
View/Download from: UTS OPUS
Meaningfully managing the relationship between representations and the entities they represent remains a challenge in robotics known as grounding. Useful insights can be found by approaching robotic systems development specifically with the grounding and symbol grounding problem in mind. In particular, Semantic Web technologies turn out to be not merely applicable to web-based software agents, but can also provide a powerful extension to existing proposals for grounded robotic systems development. Given the interoperability and openness of the Semantic Web, such technologies can increase the ability for a robot to introspect, communicate and be inspected - benefits that ultimately lead to more grounded systems with open-ended intelligent behaviour.
Johnston, B.G. & Williams, M. 2007, 'A Generic Framework for Approximate Simulation in Commonsense Reasoning Systems', International Symposium on Logical Formalizations of Commonsense R, Symposium on Logical Formalizations of Commonsense Reasoning, AAAI Press, Stanford University, USA, pp. 71-76.
View/Download from: UTS OPUS
This paper introduces the Slick architecture and outlines how it may be applied to solve the well known Egg-Cracking Problem. In contrast to other solutions to this problem that are based on formal logics, the Slick architecture is based on general- purpose and low-resolution quantitative simulations. On this benchmark problem, the Slick architecture offers greater elaboration tolerance and allows for faster elicitation of more general axioms. "This paper was selected by a process of anonymous peer reviewing for presentation at COMMONSENSE 2007" - first page of http://www.ucl.ac.uk/commonsense07/papers/johnston-and-williams.pdf "All submissions will be reviewed by the program committee listed at www.ucl.ac.uk/commonsense07/committee <http://www.ucl.ac.uk/commonsense07/committee/>, and notification of acceptance will be given by November 24, 2006." - from CFP at http://www.ucl.ac.uk/commonsense07/cfp/

Journal articles

Vitale, J., Williams, M.-.A., Johnston, B. & Boccignone, G. 2014, 'Affective facial expression processing via simulation: A probabilistic model', Biologically Inspired Cognitive Architectures, vol. 10, pp. 30-41.
View/Download from: UTS OPUS or Publisher's site
Understanding the mental state of other people is an important skill for intelligent agents and robots to operate within social environments. However, the mental processes involved in `mind-reading' are complex. One explanation of such processes is Simulation Theory - it is supported by a large body of neuropsychological research. Yet, determining the best computational model or theory to use in simulation-style emotion detection, is far from being understood. In this work, we use Simulation Theory and neuroscience findings on Mirror-Neuron Systems as the basis for a novel computational model, as a way to handle affective facial expressions. The model is based on a probabilistic mapping of observations from multiple identities onto a single fixed identity (`internal transcoding of external stimuli'), and then onto a latent space (`phenomenological response'). Together with the proposed architecture we present some promising preliminary results
Novianto, R., Johnston, B.G. & Williams, M. 2013, 'Habituation and sensitisation learning in ASMO cognitive architecture', Lecture Notes in Computer Science, vol. 8239, pp. 249-259.
View/Download from: UTS OPUS or Publisher's site
As social robots are designed to interact with humans in unstructured environments, they need to be aware of their surroundings, focus on significant events and ignore insignificant events in their environments. Humans have demonstrated a good example of adaptation to habituate and sensitise to significant and insignificant events respectively. Based on the inspiration of human habituation and sensitisation, we develop novel habituation and sensitisation mechanisms and include these mechanisms in ASMO cognitive architecture. The capability of these mechanisms is demonstrated in the `Smokey robot companion experiment. Results show that Smokey can be aware of their surroundings, focus on significant events and ignore insignificant events. ASMOs habituation and sensitisation mechanisms can be used in robots to adapt to the environment. It can also be used to modify the interaction of components in a cognitive architecture in order to improve agents or robots performances.