Pierce, K., Pepler, D. J., Craig, S. G. and Jenkin, M. Considerations for developing robot-assisted crisis de-escalation practices. Applied Sciences, 13: 4337.
Robots are increasingly entering the social sphere and taking on more sophisticated roles. One application for which robots are already being deployed is in civilian security tasks, in which robots augment security and police forces. In this domain, robots will encounter individuals in crisis who may pose a threat to themselves, others, or personal property. In such interactions with human police and security officers, a key goal is to de-escalate the situation to resolve the interaction. This paper considers the task of utilizing mobile robots in de-escalation tasks, using the mechanisms developed for de-escalation in human–human interactions. What strategies should a robot follow in order to leverage existing de-escalation approaches? Given these strategies, what sensing and interaction capabilities should a robot be capable of in order to engage in de-escalation tasks with humans? First, we discuss the current understanding of de-escalation with individuals in crisis and present a working model of the de-escalation process and strategies. Next, we review the capabilities that an autonomous agent should demonstrate to be able to apply such strategies in robot-mediated crisis de-escalation. Finally, we explore data-driven approaches to training robots in de-escalation and the next steps in moving the field forward.
Wu, D., Li, J., Ferini, A., Xu, Y. T., Jenkin, M., Jang, S, Liu, X. and Dudek, G. Reinforcement learning for communication load balancing: approaches and challenges. Front. Compute. Sci. Sec. Networks and Communications. Vol. 5.
The amount of cellular communication network traffic has increased dramatically in recent years, and this increase has led to a demand for enhanced network performance. Communication load balancing aims to balance the load across available network resources and thus improve the quality of service for network users. Most existing load balancing algorithms are manually designed and tuned rule-based methods where near-optimality is almost impossible to achieve. Furthermore, rule-based methods are difficult to adapt to quickly changing traffic patterns in real-world environments. Reinforcement learning (RL) algorithms, especially deep reinforcement learning algorithms, have achieved impressive successes in many application domains and offer the potential of good adaptabiity to dynamic changes in network load patterns. This survey presents a systematic overview of RL-based communication load-balancing methods and discusses related challenges and opportunities. We first provide an introduction to the load balancing problem and to RL from fundamental concepts to advanced models. Then, we review RL approaches that address emerging communication load balancing issues important to next generation networks, such as 5G and beyond. Finally, we highlight important challenges, open issues, and future research directions for applying RL for communication load balancing.
Bury, N. A., Jenkin, M., Allison, R. S., Herpers, R., and Harris, L. R. Vection underwater illustrates the limitations of neutral buoyancy as a microgravity analog. npj Microgravity 9 (1), 42.
Neutral buoyancy has been used as an analog for microgravity from the earliest days of human spaceflight. Compared to other options on Earth, neutral buoyancy is relatively inexpensive and presents little danger to astronauts while simulating some aspects of microgravity. Neutral buoyancy removes somatosensory cues to the direction of gravity but leaves vestibular cues intact. Removal of both somatosensory and direction of gravity cues while floating in microgravity or using virtual reality to establish conflicts between them has been shown to affect the perception of distance traveled in response to visual motion (vection) and the perception of distance. Does removal of somatosensory cues alone by neutral buoyancy similarly impact these perceptions? During neutral buoyancy we found no significant difference in either perceived distance traveled nor perceived size relative to Earth-normal conditions. This contrasts with differences in linear vection reported between short- and long-duration microgravity and Earth-normal conditions. These results indicate that neutral buoyancy is not an effective analog for microgravity for these perceptual effects.
Jenkin, H., Jenkin, M., Harris, L. R. and Herpers, R. Neutral buoyancy and the static perception of upright. npj Microgravity, 9: 52.
The perceptual upright results from the multisensory integration of the directions indicated by vision and gravity as well as a prior assumption that upright is towards the head. The direction of gravity is signalled by multiple cues, the predominant of which are the otoliths of the vestibular system and somatosensory information from contact with the support surface. Here, we used neutral buoyancy to remove somatosensory information while retaining vestibular cues, thus “splitting the gravity vector” leaving only the vestibular component. In this way, neutral buoyancy can be used as a microgravity analogue. We assessed spatial orientation using the oriented character recognition test (OChaRT, which yields the perceptual upright, PU) under both neutrally buoyant and terrestrial conditions. The effect of visual cues to upright (the visual effect) was reduced under neutral buoyancy compared to on land but the influence of gravity was unaffected. We found no significant change in the relative weighting of vision, gravity, or body cues, in contrast to results found both in long-duration microgravity and during head-down bed rest. These results indicate a relatively minor role for somatosensation in determining the perceptual upright in the presence of vestibular cues. Short-duration neutral buoyancy is a weak analogue for microgravity exposure in terms of its perceptual consequences compared to long-duration head-down bed rest.
Altarawneh, E., Agrawal, A., Jenkin, M. and Papagelis, M. Conversation Derailment Forecasting with Graph Convolutional Networks. Workshop on Online Abuse and Harms. Held in conjunction with ACL, Toronto.
Online conversations are particularly susceptible to derailment, which can manifest itself in the form of toxic communication patterns like disrespectful comments or verbal abuse. Fore-casting conversation derailment predicts signs of derailment in advance enabling proactive moderation of conversations. Current state-of-the-art approaches to address this problem relyon sequence models that treat dialogues as text streams. We propose a novel model based on a graph convolutional neural network that considers dialogue user dynamics and the influence of public perception on conversation utterances.Through empirical evaluation, we show that our model effectively captures conversation dynamics and outperforms the state-of-the-art models on the CGA and CMV benchmark datasets by1.5% and 1.7%, respectively.
Rivkin, D., Dudek, G., Kakodkar, N., Meer, D., Limoyo, O., Jenkin, M., Liu, X. and Hogan, F. ANSEL Photobot: A robot event protographer with semantic intelligence. Proc. IEEE International Conference on Robotics and Automation (ICRA). pg. 8262-8268, London, UK.
Our work examines the way in which large language models can be used for robotic planning and sampling in the context of automated photographic documentation. Specifically, we illustrate how to produce a photo-taking robot with an exceptional level of semantic awareness by leveraging recent advances in general purpose language (LM) and vision-language (VLM) models. Given a high-level description of an event we use an LM to generate a natural-language list of photo descriptions that one would expect a photographer to capture at the event. We then use a VLM to identify the best matches to these descriptions in the robot's video stream. The photo portfolios generated by our method are consistently rated as more appropriate to the event by human evaluators than those generated by existing methods.
Wu, D., Xu, Y. T., Jenkin, M., Jang, S., Hossain, E., Liu, X. and Dudek, G. Energy Saving in Cellular Wireless Networks via Transfer Deep Reinforcement Learning. Proc. IEEE Global Communications Conference, Kuala Lumpur, Malaysia.
With the increasing use of data-intensive mobile applications and the number of mobile users, the demand for wireless data services has been increasing exponentially in recent years. In order to address this demand, a large number of new cellular base stations are being deployed around the world, leading to a significant increase in energy consumption and greenhouse gas emission. Consequently, energy consumption has emerged as a key concern in the fifth-generation (5G) network era and beyond. Reinforcement learning (RL), which aims to learn a control policy via interacting with the environment, has been shown to be effective in addressing network optimization problems. However, for reinforcement learning, especially deep reinforcement learning, a large number of interactions with the environment are required. This often limits its applicability in the real world. In this work, to better deal with dynamic traffic scenarios and improve real-world applicability, we propose a transfer deep reinforcement learning framework for energy optimization in cellular communication networks. Specifically, we first pre-train a set of RL-based energy-saving policies on source base stations and then transfer the most suitable policy to the given target base station in an unsupervised learning manner. Experimental results demonstrate that base station energy consumption can be reduced significantly using this approach.
Wu, D., Xu, Y. T., Li, J., Jenkin, M., Hossain, E., Jang, S., Xin, Y., Zhang, J., Liu, X. and Dudek, G. Learning to Adapt: Communication Load Balancing via Adaptive Deep Reinforcement Learning .Proc. IEEE Global Communications Conference, Kuala Lumpur, Malaysia.
The amount of wireless communication traffic has been increasing very fast in recent years. The allocation of mobile devices among different network cells, also known as load balancing, is critical to network performance. Reinforcement learning (RL) has shown to be effective for communication load balancing and can achieve better performance than currently used rule-based methods, especially when the traffic load changes quickly. However, RL-based methods usually need to interact with the environment for a large number of time steps to learn an effective policy and can be difficult to tune. In this work, we aim to improve the data efficiency of RL-based solutions to make them more suitable for real-world applications such as wireless network load balancing. Specifically, we propose a two-stage deep RL-based solution for wireless network load balancing. In the first stage, communication load balancing is formulated as a Markov decision process, and a set of good initialization values for control actions are selected. These initialization values are chosen using some cost-efficient approach to center the training of the RL agent. In the second stage, a deep RL-based agent is trained to find offsets from the first stage that optimize the load balancing problem. Experimental evaluation of a set of dynamic traffic scenarios showcases the effectiveness of the proposed method.
Tarawneh, E., Rousseau, J.-J., Craig, S. G., Chandola, D., Khan, W., Faizi, A. and Jenkin, M. An Infrastructure for Studying the Role of Sentiment in Human-Robot Interaction. ICPR 2022 International Workshops and Challenges. ICPR 2022. Lecture Notes in Computer Science, vol 13646. Springer, Cham.
The need for social robot systems has become even more critical as a result of the ongoing pandemic. Labour shortages in the services sector and public health concerns around infection transmission combine to favour the deployment of autonomous systems in a number of traditional roles including server robots in restaurants, companion robots in long-term care homes and security robots in public spaces, to identify but a few examples. To be successful, social robots must communicate with a wide range of individuals under a wide range of different scenarios. Understanding and reacting to the sentiment being expressed by an individual is key in human-human interaction, especially in critical situations that require de-escalation. This paper takes as a starting point that user sentiment is also critical for the successful deployment of social robot systems. Although much can be learned from experiments performed in simulation, real-world experiments in the development of sentiment-aware social robots requires an infrastructure upon which to explore questions related to the role of sentiment in social robotics. This includes the development of an appropriate robot morphology and user/robot interface. This paper reports early results in the development of sentiment and display technologies as part of the development of a sentiment-informed social robot named Sentrybot, an autonomous robot intended for deployment in the security domain.
Bansal, A., Mikal, G., Surya, S., Larris, L. R., Jenkin, M. On performing vestibular damage assessment and therapy using virtual reality: lessons learned. Proc. IEEE GEM 2023, Bridgetown, Barbados, 2023 (to appear).
Virtual and augmented reality-based devices have been proposed for a range of assessment and treatment tasks, but how well are they accepted by clinicians and their patients? To investigate this question a prototype VR-based tool was developed for vestibular damage assessment and treatment. Designed to be used primarily within an in-person clinical setting, this tool was developed with the long-term goal of also supporting in-home independent and supervised treatment. Mock treatment and assessment sessions were held with non-clinical patients and the operational and patient experiences evaluated qualitatively through post-session questionnaires. Participants found the process engaging although there were concerns over hygiene, especially in light of the COVID pandemic. Clinicians felt that a VR- or AR-based approach could be effective, especially if it engaged patients in supervised, at-home exercises.
Sultana, A. and Jenkin, M. Stereo video camera calibration in the wild. Proc. ICINCO, Rome, Italy, 2023.
Although a number of robust stereo camera calibration algorithms exist in the literature, a common assumption of these algorithms is a representative set of calibration images containing a calibration target of known geometry. For stereo-video applications, it is a common practice to obtain a large number of stereo image pairs for the stereo calibration process. How should an optimal set of stereo-video calibration images be chosen when controlled camera positioning is difficult or impossible? Here we demonstrate how a RANdom SAmple Consensus (RanSaC)-based approach can be used to choose the appropriate calibration image set for improved stereo camera calibration. This paper describes the performance of RanSaC-like approach which is compared against a random frames selection approach. The performance metric is measured through mean calibration reprojection error. Evaluation on real world stereo video calibration data-sets collected in the underwater environment illustrates the effectiveness of the proposed approach.
Wu, D., Xu, Y. T., Chen, X., Wang, J., Jenkin, M., Li, H., Dudek, G. L. and Lku, X. Short-term load forecasting. US Patent US11847591B2.
A method, computer program, and computer system are provided for load forecasting. Datasets corresponding to source machine learning models and a target domain base model are identified. A set of forecasting models corresponding to the identified datasets are learned. An ensemble model is determined from the learned set of forecasting models based on gradient boosting. An available resource is allocated based on the ensemble model.
