Swarm systems consist of large numbers of robots that collaborate autonomously. With an appropria... more Swarm systems consist of large numbers of robots that collaborate autonomously. With an appropriate level of human control, swarm systems could be applied in a variety of contexts ranging from search-and-rescue situations to Cyber defence. The two decision making cycles of swarms and humans operate on two different timescales , where the former is normally orders of magnitude faster than the latter. Closing the loop at the intersection of these two cycles will create fast and adaptive humanswarm teaming networks. This paper brings desperate pieces of the ground work in this research area together to review this multidisciplinary literature. We conclude with a framework to synthesize the findings and summarize the multi-modal indicators needed for closed-loop human-swarm adaptive systems. Index Terms Human-swarm teaming (HST), human-swarm interaction (HSI), adaptive systems, automation I. INTRODUCTION Human-machine teaming (HMT) involves concurrent interactions among humans and machines. A swarm is a group of distributed machines able to self-organize and generate group-level emergent behaviors from decentralized local interactions. Human-swarm teaming (HST) extends HMT to interactions with a swarm. We will reserve the concept of Human-swarm interaction (HSI) to situations where emphasis is required for the interaction dimension, while HMT will be used when emphasis is required for the teaming dimension, and HST where the context necessitates both swarm and teaming interactions. HMT operates within a context defined by a mission. For clarity, a mission is defined by a set of objectives to be pursued by both humans and machines. Each of these objectives can be optimized through the completion of a number of tasks, which can be recursively divided into sub-tasks, and potentially could be adaptively determined while pursuing mission's objectives, while allowing for the team to negotiate plans adaptively and through their interactions. Within the context of this paper, we will assume that the overall mission has fixed and definitive objectives. Examples of these fixed objectives is to save the maximum number of lives in a search and rescue scenario or to maximize the area coverage by a swarm of air vehicles surveying a mine site. These objectives are on the HMT level. Dyer [1] offers the basic ontological constituents of a 'team' as "social members", "task interdependency", and "shared goals". Subsequent literature [2], [3] added "adaptive interaction" and "commitment" from different members in the team with distinctive skills towards performance improvement and accountability. Teaming relies on teamwork skills such as clarifying interdependencies, establishing trust, and finding out means for coordination [4]. Castellan [5] discussed the functional requirements for team members as clearly defined roles and responsibilities, task-related knowledge, and interdependent connection between one another. These three dimensions of a team can be used to distinguish teams from swarms in which members are homogeneous regarding expertise, roles, and responsibilities. The dynamic concept of teaming involves coordination and collaboration activities with flexible team structures. Bringing together the concept of "teaming" in HST raises a number of scientific challenges. Some of the challenges rest on the design of appropriate artificial intelligence algorithms to allow the swarm to be smart enough to collaborate with the human. Some challenges are epistemological in nature and call for a better understanding of distributed cognition and the form of distributed situation awareness within a swarm. This paper focuses, however, on a third form of challenges; the bidirectional communication that needs to take place between the humans and the swarms and the artificial intelligence agents that need to adapt and orchestrate this interaction. This third challenge sits at the core of the first two. Without bidirectional communication, the human-swarm teams will fall short in their abilities to effectively and efficiently collaborate. Without smart agents for bi-directional communication, the swarm and humans will find it difficult to collaborate and/or coordinate actions. Even in the simplest HST systems, a basic form of these agents are needed and may take different forms, from a pre-programmed graphical user interface to human-friendly natural communication mediums such as voice and gesture. In section II we discuss the concept of HMT and its properties, then we review possible autonomy configurations for HST, discussing risks and remedies. In section III, we distill five groups of indicators that are needed at the human-swarm interface,
There is an accumulating evidence that distracted driving is a leading cause of vehicle crashes a... more There is an accumulating evidence that distracted driving is a leading cause of vehicle crashes and accidents. In order to support safe driving, numerous methods of detecting distraction have been proposed, which are empirically focused on certain driving contexts and gaze behavior. This paper aims at illustrating a method for the non-intrusive and real-time detection of visual distraction based on vehicle dynamics data and environmental data, without using eye-tracker information. Experiments are carried out in the context of the automotive domain of the European project Holides, which addresses development and qualification of Adaptive Cooperative Human-Machine Systems, and is co-funded by ARTEMIS Joint Undertaking and Italian University, Educational and Research Department. The collected data are analised by a single layer feedforward neural network trained through pseudo-inversion methods, characterized by direct determination of output weights given randomly set input weights and biases. One main feature of our work is the convenient setting of input weights by the so called random projections: the presence of a great number of null elements in the involved matrices makes especially parsimonious the use at run time of the trained network. Moreover, we use a genetic approach to better explore the input weights network space. The obtained results show both better performance with respect to classical methods and effective and parsimonious use of memory resources.
Shallow Network Training With Dynamic Sample Weights Decay - a Potential Function Approximator for Reinforcement Learning
Neural Networks are commonly used as function approximators in Reinforcement Learning, and the Ex... more Neural Networks are commonly used as function approximators in Reinforcement Learning, and the Extreme Learning Machine is one of the best algorithms to quickly train a shallow network. The online and sequential version OS-ELM could be a great candidate to quickly train a network to be a function approximator for Reinforcement Learning, but due to its non-forgetting properties it is actually not suitable for direct use with value estimations that improve in accuracy over time. This paper presents an alternative Neural Network training algorithm based on OS-ELM, which is able to perform learning online while dynamically modifying the weights of previously learned samples in order to decrease the importance of old samples learned over time. A mathematical derivation of the formulas used is presented, along with results of experiments showing equivalence of our algorithm to ELM when learning classic datasets and the advantage provided when dealing with Reinforcement Learning data.
Swarm systems consist of large numbers of agents that collaborate autonomously. With an appropria... more Swarm systems consist of large numbers of agents that collaborate autonomously. With an appropriate level of human control, swarm systems could be applied in a variety of contexts ranging from urban search and rescue situations to cyber defence. However, the successful deployment of the swarm in such applications is conditioned by the effective coupling between human and swarm. While adaptive autonomy promises to provide enhanced performance in human-machine interaction, distinct factors must be considered for its implementation within human-swarm interaction. This paper reviews the multidisciplinary literature on different aspects contributing to the facilitation of adaptive autonomy in human-swarm interaction. Specifically, five aspects that are necessary for an adaptive agent to operate properly are considered and discussed, including mission objectives, interaction, mission complexity, automation levels, and human states. We distill the corresponding indicators in each of the fi...
2019 IEEE Symposium Series on Computational Intelligence (SSCI)
Neural Networks are commonly used as function approximators in Reinforcement Learning, and the Ex... more Neural Networks are commonly used as function approximators in Reinforcement Learning, and the Extreme Learning Machine is one of the best algorithms to quickly train a shallow network. The online and sequential version OS-ELM could be a great candidate to quickly train a network to be a function approximator for Reinforcement Learning, but due to its non-forgetting properties it is actually not suitable for direct use with value estimations that improve in accuracy over time. This paper presents an alternative Neural Network training algorithm based on OS-ELM, which is able to perform learning online while dynamically modifying the weights of previously learned samples in order to decrease the importance of old samples learned over time. A mathematical derivation of the formulas used is presented, along with results of experiments showing equivalence of our algorithm to ELM when learning classic datasets and the advantage provided when dealing with Reinforcement Learning data.
Abstract. In this paper we present a cognitively inspired system for the representation of concep... more Abstract. In this paper we present a cognitively inspired system for the representation of conceptual information in an ontology-based environment. It builds on the heterogeneous notion of concepts in Cognitive Science and on the so-called dual process theories of reasoning and rationality, and it provides a twofold view on the same artificial concept, combining a classical symbolic component (grounded on a formal ontology) with a typicality-based one (grounded on the conceptual spaces framework). The implemented system has been tested in a pilot experimentation regarding the classification task of linguistic stimuli. The results show that this modeling solution extends the representational and reasoning “conceptual ” capabilities of standard ontology-based systems. 1
Swarm systems consist of large numbers of robots that collaborate autonomously. With an appropria... more Swarm systems consist of large numbers of robots that collaborate autonomously. With an appropriate level of human control, swarm systems could be applied in a variety of contexts ranging from search-and-rescue situations to Cyber defence. The two decision making cycles of swarms and humans operate on two different time-scales, where the former is normally orders of magnitude faster than the latter. Closing the loop at the intersection of these two cycles will create fast and adaptive human-swarm teaming networks. This paper brings desperate pieces of the ground work in this research area together to review this multidisciplinary literature. We conclude with a framework to synthesize the findings and summarize the multi-modal indicators needed for closed-loop human-swarm adaptive systems.
In this paper we present a cognitively inspired system for the representation of conceptual infor... more In this paper we present a cognitively inspired system for the representation of conceptual information in an ontology-based environment. It builds on the heterogeneous notion of concepts in Cognitive Science and on the so-called dual process theories of reasoning and rationality, and it provides a twofold view on the same artificial concept, combining a classical symbolic component (grounded on a formal ontology) with a typicality-based one (grounded on the conceptual spaces framework). The implemented system has been tested in a pilot experimentation regarding the classification task of linguistic stimuli. The results show that this modeling solution extends the representational and reasoning “conceptual” capabilities of standard ontology-based systems.
There is an accumulating evidence that distracted driving is a leading cause of vehicle crashes a... more There is an accumulating evidence that distracted driving is a leading cause of vehicle crashes and accidents. In order to support safe driving, numerous methods of detecting distraction have been proposed, which are empirically focused on certain driving contexts and gaze behaviour. This paper aims at illustrating a method for the non-intrusive and real-time detection of visual distraction based on vehicle dynamics data and environmental data, without using eye-tracker information. Experiments are carried out in the context of the automotive domain of the European project Holides, which addresses development and qualification of adaptive cooperative human–machine systems, and is co-funded by ARTEMIS Joint Undertaking and Italian University, Educational and Research Department. The collected data are analysed by a single-layer feedforward neural network trained through pseudo-inversion methods, characterized by direct determination of output weights given randomly set input weights ...
Machine Learning, Optimization, and Big Data, 2016
The proposed model represents an original approach to general game playing, and aims at creating ... more The proposed model represents an original approach to general game playing, and aims at creating a player able to develop a strategy using as few requirements as possible, in order to achieve the maximum generality. The main idea is to modify the known minimax search algorithm removing its task-specific component, namely the heuristic function: this is replaced by a neural network trained to evaluate the game states using results from previous simulated matches. A method for simulating matches and extracting training examples from them is also proposed, completing the automatic procedure for the setup and improvement of the model. Part of the algorithm for extracting training examples is the Backward Iterative Deepening Search, a new original search algorithm which aims at finding, in a limited time, a high number of leaves along with their common ancestors.
In this paper we present a cognitively inspired system for the representation of conceptual infor... more In this paper we present a cognitively inspired system for the representation of conceptual information in an ontology-based envi- ronment. It builds on the heterogeneous notion of concepts in Cognitive Science and on the so-called dual process theories of reasoning and ra- tionality, and it provides a twofold view on the same articial concept, combining a classical symbolic component (grounded
In this paper we present a cognitively inspired system for the representation of conceptual infor... more In this paper we present a cognitively inspired system for the representation of conceptual information in an ontology-based environment. It builds on the heterogeneous notion of concepts in Cognitive Science and on the so-called dual process theories of reasoning and rationality, and it provides a twofold view on the same artificial concept, combining a classical symbolic component (grounded on a formal ontology) with a typicality-based one (grounded on the conceptual spaces framework). The implemented system has been tested in a pilot experimentation regarding the classification task of linguistic stimuli. The results show that this modeling solution extends the representational and reasoning "conceptual" capabilities of standard ontology-based systems.
Uploads
Papers by Leo Ghignone