Search or filter publications

Filter by type:

Filter by publication type

Filter by year:

to

Results

  • Showing results for:
  • Reset all filters

Search results

  • Conference paper
    Shilov I, Le Cadre H, Bušić A, Sanjab A, Pinson Pet al., 2025,

    Forecast Trading as a Means to Reach Social Optimum on a Peer-to-Peer Market

    , Pages: 121-130, ISSN: 0302-9743

    This paper investigates the coupling between a peer-to-peer (P2P) electricity market and a forecast market to alleviate the uncertainty faced by prosumers regarding their renewable energy sources (RES) generation. The work generalizes the analysis from Gaussian-distributed RES production to arbitrary distributions. The P2P trading is modeled as a generalized Nash equilibrium problem, where prosumers trade energy in a decentralized manner. Each agent has the option to purchase a forecast on the forecast market before trading on the electricity market. We establish conditions on arbitrary probability density functions (pdfs) under which the prosumers have incentives to purchase forecasts on the forecast market. Connected with the previous results, this allows us to prove the economic efficiency of the P2P electricity market, i.e., that a social optimum can be reached among the prosumers.
  • Journal article
    Qian Q, Wang Y, Boyle D, 2025,

    Adaptive Probabilistic Planning for the Uncertain and Dynamic Orienteering Problem

    , IEEE Internet of Things Journal

    The Orienteering Problem (OP) is a well-studied routing problem that has been extended to incorporate uncertainties, reflecting stochastic or dynamic travel costs, prize-collection costs, and prizes. Existing approaches may, however, be inefficient in real-world applications due to insufficient modeling knowledge and initially unknowable parameters in online scenarios. Thus, we propose the Uncertain and Dynamic Orienteering Problem (UDOP), modeling travel costs as distributions with unknown and time-variant parameters. UDOP also associates uncertain travel costs with dynamic prizes and prize-collection costs for its objective and budget constraints. To address UDOP, we develop an ADaptive Approach for Probabilistic paThs, ADAPT, iteratively performing 'execution' and 'online planning' based on an initial 'offline' solution. The execution phase updates the system status and records online cost observations. The online planner employs a Bayesian approach to adaptively estimate power consumption and optimize path sequence based on safety beliefs. We evaluate ADAPT in a practical Unmanned Aerial Vehicle (UAV) charging scheduling problem for Wireless Rechargeable Sensor Networks. The UAV must optimize its path to recharge sensor nodes efficiently while managing its energy under uncertain conditions. ADAPT maintains comparable solution quality and computation time while offering superior robustness. Extensive simulations show that ADAPT achieves a 100% Mission Success Rate (MSR) across all tested scenarios, outperforming comparable heuristic-based and frequentist approaches that fail up to 70% (under challenging conditions) and averaging 67% MSR, respectively. This work advances the field of OP with uncertainties, offering a reliable and efficient approach for real-world applications in uncertain and dynamic environments.
  • Journal article
    Meyer J, Picinali L, 2025,

    On the generalization of accommodation to head-related transfer functions

    , Journal of the Acoustical Society of America, Vol: 157, Pages: 420-432, ISSN: 0001-4966

    To date, there is strong evidence indicating that humans with normal hearing can adapt to non-individual head-related transfer functions (HRTFs). However, less attention has been given to studying the generalization of this adaptation to untrained conditions. This study investigated how adaptation to one set of HRTFs can generalize to another set of HRTFs. Participants were divided into two groups and trained to localize a speech stimulus reproduced binaurally using either individual or non-individual HRTFs. Training led to an improved localization performance with the trained HRTFs for both groups of participants. Results also showed that there was no difference in the localization performance improvement between the trained and untrained HRTFs for both groups, indicating a generalization of adaptation to HRTFs. The findings did not allow to precisely determine which type of learning (procedural or perceptual) primarily contributed to the generalization, thus highlighting the potential need to expose participants to longer training protocols.
  • Journal article
    Chen X, Chen W, Lee D, Ge Y, Rojas N, Kormushev Pet al., 2025,

    A Backbone for Long-Horizon Robot Task Understanding

    , IEEE Robotics and Automation Letters, Vol: 10, Pages: 2048-2055

    End-to-end robotlearning, particularly for long-horizon tasks, often results in unpredictable outcomes and poor generalization. To address these challenges, we propose a novel Therblig-Based Backbone Framework (TBBF) as a fundamental structure to enhance interpretability, data efficiency, and generalization in robotic systems. TBBF utilizes expert demonstrations to enable therblig-level task decomposition, facilitate efficient action-object mapping, and generate adaptive trajectories for new scenarios. The approach consists of two stages: offline training and online testing. During the offline training stage, we developed the Meta-RGate SynerFusion (MGSF) network for accurate therblig segmentation across various tasks. In the online testing stage, after a one-shot demonstration of a new task is collected, our MGSF network extracts high-level knowledge, which is then encoded into the image using Action Registration (ActionREG). Additionally, Large Language Model (LLM)-Alignment Policy for Visual Correction (LAP-VC) is employed to ensure precise action registration, facilitating trajectory transfer in novel robot scenarios. Experimental results validate these methods, achieving 94.37% recall in therblig segmentation and success rates of 94.4% and 80% in real-world online robot testing for simple and complex scenarios, respectively.
  • Journal article
    Rizvi D, Boyle D, 2025,

    Multi-agent reinforcement learning with action masking for UAV-enabled mobile communications

    , IEEE Transactions on Machine Learning in Communications and Networking, Vol: 3, Pages: 117-132, ISSN: 2831-316X

    Unmanned Aerial Vehicles (UAVs) are increasingly used as aerial base stations to provide ad hoc communications infrastructure. Building upon prior research efforts which consider either static nodes, 2D trajectories or single UAV systems, this paper focuses on the use of multiple UAVs for providing wireless communication to mobile users in the absence of terrestrial communications infrastructure. In particular, we jointly optimize UAV 3D trajectory and NOMA power allocation to maximize system throughput. Firstly, a weighted K-means-based clustering algorithm establishes UAV-user associations at regular intervals. Then the efficacy of training a novel Shared Deep Q-Network (SDQN) with action masking is explored. Unlike training each UAV separately using DQN, the SDQN reduces training time by using the experiences of multiple UAVs instead of a single agent. We also show that SDQN can be used to train a multi-agent system with differing action spaces. Simulation results confirm that: 1) training a shared DQN outperforms a conventional DQN in terms of maximum system throughput (+20%) and training time (-10%); 2) it can converge for agents with different action spaces, yielding a 9% increase in throughput compared to Mutual DQN algorithm; and 3) combining NOMA with an SDQN architecture enables the network to achieve a better sum rate compared with existing baseline schemes.
  • Journal article
    Wu H, Lu Z, Hill S, Turner Ret al., 2025,

    Microstructure Characterisation and Modelling of Pre-Forging Solution Treatment of 7075 Aluminium Alloy Using Novel Heating Methods

    , Journal of Manufacturing and Materials Processing, Vol: 9

    This study evaluates the effectiveness of these conventional heating methods, commonly adopted in the industry with long durations (typically one hour), in comparison to newer, potentially more efficient approaches such as induction coil heating, infrared module heating, and infrared furnaces that can perform solution heat treatment in significantly shorter times (5 to 20 min). The properties of the edge and centre regions of the solution-treated billets, including the state of precipitates, grain structures, and Vickers hardness, are investigated and compared. Results have shown that the 7075 billets heated by conventional heating methods sufficiently dissolved the stable precipitates, achieving hardness ranging from 137 to 141 HV, in contrast to the benchmark unheated, as-received sample of approximately 70 HV. In the meantime, the induction coil and infrared furnace demonstrate notable effectiveness, achieving hardness between 126 and 135 HV. The average grain sizes in the centre and edge regions for all samples are measured as 3 and 8 µm, respectively. However, the impact of the grain size on the hardness is negligible compared to the impact of the precipitates. Finite element (FE) modelling comparing the slowest heating method—the electric furnace—and the fastest heating method—induction coil heating—reveals the latter could heat the billet up to 450 °C at a rate ten times faster than the electric furnace. This study highlights the potential of novel heating techniques in promoting the efficiency of heat treatment processes for 7075 aluminium alloys.
  • Journal article
    Rostami-Tabar B, Pinson P, Porter MD, 2025,

    Guest editorial: Forecasting for social good

    , INTERNATIONAL JOURNAL OF FORECASTING, Vol: 41, Pages: 1-2, ISSN: 0169-2070
  • Journal article
    Wang C, Pinson P, Wang Y, 2025,

    Seamless and Multi-Resolution Energy Forecasting

    , IEEE Transactions on Smart Grid, Vol: 16, Pages: 383-395, ISSN: 1949-3053

    Forecasting is pivotal in energy systems, by providing fundamentals for operation at different horizons and resolutions. Though energy forecasting has been widely studied for capturing temporal information, very few works concentrate on the frequency information provided by forecasts. They are consequently often limited to single-resolution applications (e.g., hourly). Here, we propose a unified energy forecasting framework based on Laplace transform in the multi-resolution context. The forecasts can be seamlessly produced at different desired resolutions without re-training or post-processing. Case studies on both energy demand and supply data show that the forecasts from our proposed method can provide accurate information in both time and frequency domains. Across the resolutions, the forecasts also demonstrate high consistency. More importantly, we explore the operational effects of our produced forecasts in the day-ahead and intra-day energy scheduling. The relationship between (i) errors in both time and frequency domains and (ii) operational value of the forecasts is analyzed. Significant operational benefits are obtained.
  • Journal article
    Wang Y, Boyle D, 2025,

    Constrained reinforcement learning using distributional representation for trustworthy quadrotor UAV tracking control

    , IEEE Transactions on Automation Science and Engineering, Vol: 22, Pages: 5877-5894, ISSN: 1545-5955

    Simultaneously accurate and reliable tracking control for quadrotors in complex dynamic environments is challenging. The chaotic nature of aerodynamics, derived from drag forces and moment variations, makes precise identification difficult. Consequently, many existing quadrotor tracking systems treat these aerodynamic effects as simple ‘disturbances’ in conventional control approaches. We propose a novel and interpretable trajectory tracker integrating a distributional Reinforcement Learning (RL) disturbance estimator for unknown aerodynamic effects with a Stochastic Model Predictive Controller (SMPC). Specifically, the proposed estimator ‘Constrained Distributional REinforced-Disturbance-estimator’ (ConsDRED) effectively identifies uncertainties between the true and estimated values of aerodynamic effects. Control parameterization employs simplified affine disturbance feedback to ensure convexity, which is seamlessly integrated with the SMPC. We theoretically guarantee that ConsDRED achieves an optimal global convergence rate, and sublinear rates if constraints are violated with certain error decreases as neural network dimensions increase. To demonstrate practicality, we show convergent training, in simulation and real-world experiments, and empirically verify that ConsDRED is less sensitive to hyperparameter settings compared with canonical constrained RL. Our system substantially improves accumulative tracking errors by at least 70%, compared with the recent art. Importantly, the proposed ConsDRED-SMPC framework balances the trade-off between pursuing high performance and obeying conservative constraints for practical implementations. Note to Practitioners —This work is motivated by challenges in training Reinforcement Learning (RL) for autonomous navigation in unmanned aerial vehicles, but its implications extend to other high-criticality applications in, for example, healthcare and financial services. The implementation of RL algo
  • Journal article
    Sadek M, Kallina E, Bohné T, Mougenot C, Calvo RA, Cave Set al., 2025,

    Challenges of responsible AI in practice: scoping review and recommended actions

    , AI and Society: the journal of human-centered systems and machine intelligence, Vol: 40, Pages: 199-215, ISSN: 0951-5666

    Responsible AI (RAI) guidelines aim to ensure that AI systems respect democratic values. While a step in the right direction, they currently fail to impact practice. Our work discusses reasons for this lack of impact and clusters them into five areas: (1) the abstract nature of RAI guidelines, (2) the problem of selecting and reconciling values, (3) the difficulty of operationalising RAI success metrics, (4) the fragmentation of the AI pipeline, and (5) the lack of internal advocacy and accountability. Afterwards, we introduce a number of approaches to RAI from a range of disciplines, exploring their potential as solutions to the identified challenges. We anchor these solutions in practice through concrete examples, bridging the gap between the theoretical considerations of RAI and on-the-ground processes that currently shape how AI systems are built. Our work considers the socio-technical nature of RAI limitations and the resulting necessity of producing socio-technical solutions.
  • Journal article
    Frade JLH, Giraldi JDME, Porat T, 2025,

    The country-of-origin effect on vaccination: a systematic literature review and research agenda

    , Management Review Quarterly, ISSN: 2198-1620

    The COVID-19 pandemic raised awareness and concerns regarding the country-of-origin of vaccines. During this period, we witnessed the emergence of a country-of-origin effect in vaccination perceptions. The country-of-origin effect is a well-documented marketing phenomenon where the origin country of a product influences consumer decisions, brand associations, and evaluations. To investigate this phenomenon, a systematic literature review was conducted using Scopus database, employing a diverse array of search terms. The review identified 52 articles that examined the country-of-origin effect on vaccination. These studies fall under different subject fields, such as Medicine and Social Science, and were published across 39 different journals, confirming the interdisciplinary nature of the topic. Moreover, the studies covered 48 countries, with some being multicultural. The results reveal the presence of a national bias, a preference for Western vaccines, distrust towards Chinese and Russian vaccines, and the impact of demographic factors, such as gender, age, and income. The national bias was observed across at least 17 countries, such as USA, China, UK, Germany, Turkey and Iran. It persists even in countries without early COVID-19 vaccine development (e.g., Brazil, Ghana, Japan, South Korea, Spain and Taiwan). The preference for Western vaccines and distrust towards Chinese and Russian vaccines was observed across diverse regions including Europe (e.g., France), Latin American (e.g., Brazil), Eastern Europe (e.g., Hungary), the Middle East (e.g., Turkey, Israel), and Asia (e.g., Japan). The authors discuss potential underlying reasons, implications for policy makers and health management, and propose a comprehensive agenda for future research, including the role of politics, media, endorsements, and other vaccines and medications.
  • Journal article
    Arana-Catania M, Sonee A, Khan AM, Fatehi K, Tang Y, Jin B, Soligo A, Boyle D, Calinescu R, Yadav P, Ahmadi H, Tsourdos A, Guo W, Russo Aet al., 2025,

    Explainable Reinforcement and Causal Learning for Improving Trust to 6G Stakeholders

    , IEEE Open Journal of the Communications Society

    Future telecommunications will increasingly integrate AI capabilities into network infrastructures to deliver seamless and harmonized services closer to end-users. However, this progress also raises significant trust and safety concerns. The machine learning systems orchestrating these advanced services will widely rely on deep reinforcement learning (DRL) to process multi-modal requirements datasets and make semantically modulated decisions, introducing three major challenges: (1) First, we acknowledge that most explainable AI research is stakeholder-agnostic while, in reality, the explanations must cater for diverse telecommunications stakeholders, including network service providers, legal authorities, and end users, each with unique goals and operational practices; (2) Second, DRL lacks prior models or established frameworks to guide the creation of meaningful long-term explanations of the agent’s behaviour in a goal-oriented RL task, and we introduce state-of-the-art approaches such as reward machine and sub-goal automata that can be universally represented and easily manipulated by logic programs and verifiably learned by inductive logic programming of answer set programs; (3) Third, most explainability approaches focus on correlation rather than causation, and we emphasise that understanding causal learning can further enhance 6G network optimisation. Together, in our judgement they form crucial enabling technologies for trustworthy services in 6G. This review offers a timely resource for academic researchers and industry practitioners by highlighting the methodological advancements needed for explainable DRL (X-DRL) in 6G. It identifies key stakeholder groups, maps their needs to X-DRL solutions, and presents case studies showcasing practical applications. By identifying and analysing these challenges in the context of 6G case studies, this work aims to inform future research, transform industry practices, and highlight unresolved gaps in this rapidly
  • Journal article
    Pierrot A, Pinson P, 2025,

    Data Are Missing Again-Reconstruction of Power Generation Data Using <i>k</i>-Nearest Neighbors and Spectral Graph Theory

    , WIND ENERGY, Vol: 28, ISSN: 1095-4244
  • Journal article
    , 2025,

    Why Sponge Planet? Discussions on Land-Based, Water-Driven Solutions

    , Landscape Architecture Frontiers, Vol: 0, Pages: 0-0, ISSN: 2096-336X
  • Book chapter
    Vohra S, Childs PRN, 2025,

    Interplays Between Learning Theories

    , Design Science and Innovation, Publisher: Springer Nature Singapore, Pages: 57-71, ISBN: 9789819773626
  • Journal article
    Jagtap SS, Childs PRN, Stettler MEJ, 2024,

    Conceptual design-optimisation of a subsonic hydrogen-powered long-range blended-wing-body aircraft

    , INTERNATIONAL JOURNAL OF HYDROGEN ENERGY, Vol: 96, Pages: 639-651, ISSN: 0360-3199
  • Journal article
    Riley S, Shevchuk A, George C, 2024,

    Achieving dynamic stability and electromechanical resilience for ultra-flexible battery technology

    , COMMUNICATIONS MATERIALS, Vol: 5
  • Journal article
    Jagtap SS, Childs PRN, Stettler MEJ, 2024,

    Conceptual design-optimisation of a future hydrogen-powered ultrahigh bypass ratio geared turbofan engine

    , International Journal of Hydrogen Energy, Vol: 95, Pages: 317-328, ISSN: 0360-3199

    Liquid hydrogen (LH2) is a proposed option to decarbonise long-haul aviation. LH2 aircraft (combustion-based) is expected to be lighter than Jet-A aircraft which necessitates reduction in the engine thrust requirement. Thus, the thermodynamic and energy performance of a LH2 aircraft engine, and its design and optimisation, is of significance. In a first, a conceptual design and optimisation of a future LH2 powered ultra-high bypass-ratio geared turbofan engine is conducted for reduced aircraft thrust requirement, using GasTurb 13 software and implementing future materials and component efficiencies. The thrust specific energy consumption (TSEC) of the optimised LH2 engine is 6–8% lower than Jet-A. The TSEC of LH2 engine is lower than Jet-A due to hydrogen's higher gravimetric energy density during combustion, higher specific heat of combustion products, and reduced thrust requirement. It is observed that optimised LH2 engine has 11% smaller diameter, 5.5–7.5% shorter length, 6–14% lower turbine entry temperature and 7.4–17.6% lower weight, than a Jet-A engine. The results of this work will be useful to future studies on LH2 engine and aircraft design, and LH2 aircraft emissions and contrails modelling.
  • Journal article
    Sadek M, Mougenot C, 2024,

    Challenges in Value-Sensitive AI Design: Insights from AI Practitioner Interviews

    , INTERNATIONAL JOURNAL OF HUMAN-COMPUTER INTERACTION, ISSN: 1044-7318
  • Journal article
    Liu X, Yang K, Zhang L, Wang W, Zhou S, Wu B, Xiong M, Yang S, Tan Ret al., 2024,

    A Fast Forward Prediction Framework for Energy Materials Design Based on Machine Learning Methods

    , ENERGY MATERIAL ADVANCES, Vol: 5, ISSN: 2097-1133
  • Journal article
    McPherson A, Morrison L, Davison M, Wanderley Met al., 2024,

    On mapping as a technoscientific practice in digital musical instruments

    , Journal of New Music Research, ISSN: 0929-8215

    This article provides historical context for the emergence of “mapping” as a keyconceptual metaphor in the context of digital musical instrument (DMI) design anduse. In addition to a consideration of different technical implementations, we offer a critical assessment of the tendency to over-generalise mapping as a universalmodel for both building instruments and analysing them in retrospect. This reification of mapping as a design model, as well as of the dimension spaces of soundand gesture being mapped, is read through a media-theoretical lens, drawing onrecent work from interface studies to show how mapping actively constructs ideological relationships between performers and underlying systems of musical representation. While acknowledging the practical utility of traditional formulations ofmapping in DMIs, we focus on issues arising from their over-generalisation, includingthe sometimes-misleading impression of representational stability, the suitability ofspatial metaphors, and the assumption of unidirectionality and temporal stasis. Inclosing, the article explores alternatives based on a relational approach to mappingas an “intra-active” process that is bidirectional at every step, fluid in its distinctionof categories, and more dynamic across its variegated temporalities.
  • Journal article
    Smith F, Sadek M, Wan E, Ito A, Mougenot Cet al., 2024,

    Codesigning AI with End-Users: An AI Literacy Toolkit for Nontechnical Audiences (Jul, 10.1093/iwc/iwae029, 2024)

    , INTERACTING WITH COMPUTERS, ISSN: 0953-5438
  • Journal article
    Le Penru NP, Heath BE, Dunning J, Picinali L, Ewers RM, Sethi SSet al., 2024,

    Towards using virtual acoustics for evaluating spatial ecoacoustic monitoring technologies

    , METHODS IN ECOLOGY AND EVOLUTION, ISSN: 2041-210X
  • Journal article
    Kench S, Squires I, Dahari A, Planella FB, Roberts SA, Cooper SJet al., 2024,

    Li-ion battery design through microstructural optimization using generative AI

    , MATTER, Vol: 7, ISSN: 2590-2393
  • Journal article
    Daubner S, Nestler B, 2024,

    Microstructure Characterization of Battery Materials Based on Voxelated Image Data: Computation of Active Surface Area and Tortuosity

    , JOURNAL OF THE ELECTROCHEMICAL SOCIETY, Vol: 171, ISSN: 0013-4651
  • Book
    Nanayakkara T, 2024,

    Handbook on soft robotics

    This book explains how to design and control a soft robot in understandable language. In addition, it provides a comprehensive coverage of the essential theory and techniques used in soft robotics that can be used by graduate students in soft robotics. The book covers several key areas in soft robots, ranging from design and fabrication to modelling and control. It also includes many case studies and examples. The book clearly explains mathematical concepts and uses illustrative explanation to help engineers and junior graduate students understand the physical meaning of the key concepts and approaches in soft robotics. Reading this book gives professional engineers and students a sound knowledge of soft robotics that they can take to their careers and research.
  • Journal article
    Nanayakkara T, 2024,

    Preface II

    , Handbook on Soft Robotics, Pages: xiii-xv
  • Journal article
    Nanayakkara T, 2024,

    Introduction

    , Handbook on Soft Robotics, Pages: 1-12

    Softness has a meaning relative to the forces and their frequencies an object experiences.
  • Book chapter
    Nanayakkara T, Mulvey B, Perera S, Ge Y, Yu Z, Sunilkumar Pet al., 2024,

    Soft robots as a platform to understand embodied intelligence

    , Handbook on Soft Robotics, Pages: 35-84

    This chapter addresses some of the possible missing pieces to understand the gulf between robots and biological beings in negotiating movements in a natural world. More specifically, we discuss the overarching question of understanding how passive dynamics and computing can come together to solve complex interaction problems. The elegance of physical interactions we have in daily life appears to come from a fine interplay among the dynamics and computing in the brain, the body, and the environment. However, it is difficult to map this out using human participants alone. Can we use robots as a proxy to living beings to understand the nuances of such interactions since we can control and measure details of information flow? We will group examples under three phenomena we believe are important to untangle the puzzle: Ghost circuits, kinematic tuning, and behavioral lensing on which we will elaborate further in this chapter.
  • Journal article
    Nanayakkara T, 2024,

    Preface I

    , Handbook on Soft Robotics

This data is extracted from the Web of Science and reproduced under a licence from Thomson Reuters. You may not copy or re-distribute this data in whole or in part without the written consent of the Science business of Thomson Reuters.

Request URL: http://www.imperial.ac.uk:80/respub/WEB-INF/jsp/search-t4-html.jsp Request URI: /respub/WEB-INF/jsp/search-t4-html.jsp Query String: id=1221&limit=30&resgrpMemberPubs=true&resgrpMemberPubs=true&page=2&respub-action=search.html Current Millis: 1747895704429 Current Time: Thu May 22 07:35:04 BST 2025

Contact us

Dyson School of Design Engineering
Imperial College London
25 Exhibition Road
South Kensington
London
SW7 2DB

design.engineering@imperial.ac.uk
Tel: +44 (0) 20 7594 8888

Campus Map