Article
Automation & Control Systems
Quan Nguyen, Koushil Sreenath
Summary: In this paper, we propose a novel method of optimal robust control for nonlinear dynamical robotic systems in the presence of model uncertainty. The method offers tracking stability while considering input and state-based constraints as well as safety-critical constraints.
IEEE TRANSACTIONS ON AUTOMATIC CONTROL
(2022)
Review
Anesthesiology
Joshua Biro, Maya Rucks, David M. Neyens, Sarah Coppola, James H. Abernathy, Ken R. Catchpole
Summary: This article examines the variability and inconsistencies in the definitions of patient safety-related terminology in anaesthesia. The analysis reveals that there is wide variation in the definitions of medication errors, while definitions of terms in other categories are relatively consistent. The inconsistency in terminology may affect the understanding and synthesis of anaesthesia medication safety.
BRITISH JOURNAL OF ANAESTHESIA
(2022)
Article
Robotics
Tao Jin, Jian Di, Xinghu Wang, Haibo Ji
Summary: This letter presents a novel predictive control algorithm that combines control barrier function (CBF) technique with path integral control to develop a safety-critical controller for quadrotors. An efficient sequential projection algorithm is adopted to steer unsafe sampled inputs to a feasible region limited by safety barrier certificates. Comparisons with existing methods show that the proposed controller, which provides collective thrusts and body rates (CTBR) control policies, is able to deliver more robust maneuvers in obstacle avoidance.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2023)
Article
Automation & Control Systems
Adam K. Kiss, Tamas G. Molnar, Aaron D. Ames, Gabor Orosz
Summary: This work presents a theoretical framework for the safety-critical control of time delay systems. It extends the theory of control barrier functions to systems with state delay, introducing the notion of control barrier functionals to attain formal safety guarantees in infinite dimensional state space. The proposed framework can handle multiple delays and distributed delays in both dynamics and safety conditions, providing an affine constraint on the control input for provable safety.
INTERNATIONAL JOURNAL OF ROBUST AND NONLINEAR CONTROL
(2023)
Editorial Material
Automation & Control Systems
Rodolphe Sepulchre
Summary: This issue of IEEE Control Systems includes a feature on safety filtering approaches for complex control systems and a lecture note on the Prandtl-Ishlinskii hysteresis model and its inverse compensator. The feature article provides an introductory tutorial on runtime assurance concepts and their role in ensuring safety, while the lecture note introduces the mathematical formulations of the PI and GPI hysteresis models and their analytical inverses. Both articles are valuable resources for researchers and graduate students in the field of control systems.
IEEE CONTROL SYSTEMS MAGAZINE
(2023)
Article
Medicine, General & Internal
Philippa Rees, Thomas Purchase, Emily Ball, Jillian Beggs, Francesca Gabriel, Sioned Gwyn, Stuart Hellard, Elena Jones, Isobel Joy McFadzean, Davide Paccagnella, Philippa Robb, Kathleen Walsh, Andrew Carson-Stevens
Summary: This study aims to explore the role of families, guardians, and parents in paediatric safety incidents and how this role may have changed during the pandemic. The findings will help in delivering safer care and developing harm prevention strategies across healthcare settings.
Editorial Material
Anesthesiology
David Lockey
Summary: The identification, triage, and extrication of casualties, as well as on-scene management and transport, can be complex and cause significant delays. An effective pre-hospital pathway is crucial for increasing survival chances and improving emergency response effectiveness.
BRITISH JOURNAL OF ANAESTHESIA
(2022)
Article
Automation & Control Systems
Tamas G. Molnar, Adam K. Kiss, Aaron D. Ames, Gabor Orosz
Summary: This article develops a framework for safety-critical control in dynamic environments by introducing the concept of environmental control barrier functions (ECBFs). This framework can guarantee safety even with input delay by considering the evolution of the environment during the delayed response of the system. The proposed method is demonstrated to be effective in a simple adaptive cruise control (ACC) problem and a more complex robotics application on a Segway platform.
IEEE TRANSACTIONS ON CONTROL SYSTEMS TECHNOLOGY
(2023)
Article
Pediatrics
Luise Brado, Susanne Tippmann, Daniel Schreiner, Jonas Scherer, Dorothea Plaschka, Eva Mildenberger, Andre Kidszun
Summary: Safety incidents were identified and analyzed in a German NICU over a 6-month period, with medication errors and equipment problems being the most commonly reported issues. The majority of incidents were deemed preventable, with suggestions for improvement focused on the implementation of computer-assisted tools and processes for enhancing care quality.
FRONTIERS IN PEDIATRICS
(2021)
Article
Robotics
Tamas G. Molnar, Ryan K. Cosner, Andrew W. Singletary, Wyatt Ubellacker, Aaron D. Ames
Summary: This letter presents a framework for safety-critical control of robotic systems based on safe regions in the configuration space. It synthesizes a safe velocity using the control barrier function theory without relying on a complicated dynamic model and tracks the safe velocity with a tracking controller. The proposed method provides theoretical safety guarantees and has been demonstrated to be application-agnostic.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2022)
Article
Automation & Control Systems
Sihua Zhang, Di-Hua Zhai, Yuhan Xiong, Yuanqing Xia
Summary: This article proposes a feasibility-guaranteed quadratic programming (QP) approach to address the conflict between high-order control barrier function (HOCBF) constraint and input constraint. The method first updates the parameters by adding a feasibility constraint derived from the input constraint and HOCBF constraint in the classical QP. Then the Type-2 HOCBF is investigated to effectively restrict the system within a single HOCBF at the current time step for systems with multiple HOCBF constraints. The efficacy of this approach is demonstrated through the application of obstacle avoidance in a 3-DOF robot system.
INTERNATIONAL JOURNAL OF ROBUST AND NONLINEAR CONTROL
(2023)
Article
Health Care Sciences & Services
Ke LI, Xueyan Cao, Zhiwei He, Liqun Liu
Summary: This research aimed to explore the effects of perceived knowledge gap, risk perception, past actual risk experience, and media risk experience on anxiety. The results showed that actual risk experience and media risk experience have different influence mechanisms on anxiety.
Article
Engineering, Industrial
Abdullah Alsharef, Alex Albert, S. M. Jamil Uddin, Nikhil Basavaraj Kittur, Sampada Chavan, Edward Jaselskis
Summary: Driver license examiners serve as a first line of defense against unsafe drivers, yet they themselves face high safety risks during testing. The study found that examiners experience numerous incidents, with common event types, contributing factors, injury types, and outcomes identified. Specific relationships in incident reports, such as collision with fixed objects, were also highlighted to inform future injury prevention efforts.
Article
Automation & Control Systems
Wei Xiao, Calin Belta, Christos G. Cassandras
Summary: This article addresses the problem of safety-critical control for multiagent systems with unknown dynamics in unknown environments. It demonstrates the reduction of stabilizing affine control systems to quadratic programs using control barrier functions (CBFs) and control Lyapunov functions (CLFs). The article proposes a robust framework that updates adaptive affine control dynamics based on real-time sensor measurements and error states, and reformulates the safety-critical control problem as a sequence of quadratic programs.
IEEE TRANSACTIONS ON AUTOMATIC CONTROL
(2023)
Article
Automation & Control Systems
Xiao Tan, Wenceslao Shaw Cortez, Dimos V. Dimarogonas
Summary: This article introduces a notion of high-order barrier functions that guarantee set forward invariance by checking their higher order derivatives. A singularity-free control scheme is proposed for controlled dynamical systems, ensuring safety. The article also includes a case study on rigid-body attitude dynamics.
IEEE TRANSACTIONS ON AUTOMATIC CONTROL
(2022)
Article
Computer Science, Cybernetics
Luis A. Leiva, Morteza Shiripour, Antti Oulasvirta
Summary: Aesthetics is a crucial factor in user interface design, however, the perception of pleasant design varies greatly among users. We have developed a computational model that can estimate the visual appeal of a webpage based on different user groups, and it can assist designers in creating personalized designs and evaluating webpage design prototypes quickly.
UNIVERSAL ACCESS IN THE INFORMATION SOCIETY
(2023)
Article
Computer Science, Artificial Intelligence
Luis A. Leiva, Asutosh Hota, Antti Oulasvirta
Summary: XUI is a novel method that uses computational models to provide informative descriptions of user interfaces, including an overview and detailed descriptions, enhancing understanding of the UI's purpose.
ACM TRANSACTIONS ON INTELLIGENT SYSTEMS AND TECHNOLOGY
(2023)
Article
Computer Science, Information Systems
Yi-Chi Liao, John J. Dudley, George B. Mo, Chun-Lien Cheng, Liwei Chan, Antti Oulasvirta, Per Ola Kristensson
Summary: This article examines how AI can facilitate the process of interaction design by offloading complex decision making needed by designers. It discusses the use of multi-objective Bayesian optimization to support designers in creating tactile displays for smart watches. The study presents the advantages and disadvantages of utilizing the human-AI collaboration enabled by multi-objective Bayesian optimization over conventional design practice.
IEEE PERVASIVE COMPUTING
(2023)
Article
Computer Science, Artificial Intelligence
Sebastiaan De Peuter, Antti Oulasvirta, Samuel Kaski
Summary: We need to rethink how AI can assist designers by supporting their creativity and problem-solving capabilities instead of automating their tasks. The challenge is to understand designers' goals and provide help without interrupting their workflow. AI-assisted design introduces a framework based on generative user models to infer and adapt to designers' goals, reasoning, and capabilities.
Article
Computer Science, Artificial Intelligence
Andrew Howes, Jussi P. P. Jokinen, Antti Oulasvirta
Summary: The ability to estimate human behavior state is not sufficient for building cooperative agents. Predicting how people adapt their behavior in response to an agent's actions is also necessary. We propose a new approach based on computational rationality, which combines reinforcement learning and cognitive modeling to facilitate machine understanding of humans.
Proceedings Paper
Computer Science, Cybernetics
Suyog Chandramouli, Yifan Zhu, Antti Oulasvirta
Summary: This article discusses the personalization of opaque-box image classifiers using an interactive hyperparameter tuning approach. By iteratively rating the quality of explanations for a selected set of query images, the classifier's accuracy and perceived explainability ratings are optimized using a multi-objective Bayesian optimization algorithm. The study found that adjusting hyperparameters can significantly improve the explainability ratings of queried images while minimally impacting classifier accuracy. This method has the potential to be used for joint optimization of any machine learning objective and any human-centric objective.
2023 PROCEEDINGS OF THE 31ST ACM CONFERENCE ON USER MODELING, ADAPTATION AND PERSONALIZATION, UMAP 2023
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Yueyang Wang, Aravinda Ramakrishnan Srinivasan, Jussi P. P. Jokinen, Antti Oulasvirta, Gustav Markkula
Summary: Understanding the interaction between different road users is crucial for road safety and automated vehicles. Existing mathematical models on this topic have been mostly proposed based on cognitive or machine learning approaches. However, current cognitive models fail to simulate road user trajectories in general scenarios, while machine learning models lack a focus on the mechanisms generating behavior and may not capture important human-like behaviors. In this study, we develop a computational rationality model using deep reinforcement learning to capture human pedestrian crossing decisions, considering the limited human visual system. Our results demonstrate that the proposed cognitive-reinforcement learning model replicates human-like patterns of gap acceptance and crossing initiation time, providing new insights into road user behavior.
2023 IEEE INTELLIGENT VEHICLES SYMPOSIUM, IV
(2023)
Proceedings Paper
Computer Science, Information Systems
Aini Putkonen, Aurelien Nioche, Markku Laine, Crista Kuuramo, Antti Oulasvirta
Summary: This study investigates the distribution of visual attention among content items when browsing news. The results suggest that visual attention in browsing is fragmented, and influenced by the number, properties, and composition of the items visible on the viewport.
ADVANCES IN INFORMATION RETRIEVAL, ECIR 2023, PT II
(2023)
Proceedings Paper
Computer Science, Information Systems
Hee-Seung Moon, Antti Oulasvirta, Byungjoo Lee
Summary: There has been significant progress in simulation models for predicting human behavior, but determining parameter values and the time-consuming parameter inference remain challenges. This study explores amortized inference as a solution to dramatically reduce inference time and demonstrates its effectiveness in analyzing large-scale datasets for human-computer interaction tasks. The study also highlights emerging opportunities and challenges in applying amortized inference in HCI.
PROCEEDINGS OF THE 2023 CHI CONFERENCE ON HUMAN FACTORS IN COMPUTING SYSTEMS (CHI 2023)
(2023)
Proceedings Paper
Computer Science, Information Systems
Yue Jiang, Luis A. Leiva, Paul R. B. Houssel, Hamed R. Tavakoli, Julia Kylmala, Antti Oulasvirta
Summary: This article introduces a large eye-tracking-based dataset UEyes and its analysis results. The authors compare and analyze the influences of factors such as color, location, and gaze direction on four major UI types: webpage, desktop UI, mobile UI, and poster. They also propose improvements for predictive models to better capture typical tendencies across UI types.
PROCEEDINGS OF THE 2023 CHI CONFERENCE ON HUMAN FACTORS IN COMPUTING SYSTEMS, CHI 2023
(2023)
Proceedings Paper
Computer Science, Information Systems
Zhi Li, Yu-Jung Ko, Aini Putkonen, Shirin Feiz, Vikas Ashok, Iv Ramakrishnan, Antti Oulasvirta, Xiaojun Bi
Summary: This paper proposes a computational model that simulates blind users' menu selection behavior. The model takes into account the impact of long-term memory on users' selection behavior and is validated against empirical study data.
PROCEEDINGS OF THE 2023 CHI CONFERENCE ON HUMAN FACTORS IN COMPUTING SYSTEMS, CHI 2023
(2023)
Article
Computer Science, Information Systems
Tapani Rinta-Kahila, Esko Penttinen, Antti Salovaara, Wael Soliman, Joona Ruissalo
Summary: Cognitive automation, powered by advanced intelligent technologies, enables organizations to automate more knowledge work tasks. However, this also leads to the erosion of human skills and expertise, as workers become less mindful in tasks reliant on automation. This study examines the dynamics behind such skill erosion, specifically in an accounting firm, and highlights the role of automation reliance and complacency in weakening workers' mindfulness and causing skill erosion.
JOURNAL OF THE ASSOCIATION FOR INFORMATION SYSTEMS
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Aini Putkonen, Aurelien Nioche, Ville Tanskanen, Arto Klami, Antti Oulasvirta
Summary: Theory-based models have interpretable parameters, but their inference can be challenging in real-world applications. This paper proposes a technique to assess the applicability of naturalistic datasets and demonstrates its use in two decision-making models under risk.
PROCEEDINGS OF THE 30TH ACM CONFERENCE ON USER MODELING, ADAPTATION AND PERSONALIZATION, UMAP 2022
(2022)
Proceedings Paper
Antti Oulasvirta, Jussi P. P. Jokinen, Andrew Howes
Summary: This paper examines how people interact with computers and proposes using computational rationality as a theoretical framework to find answers. Research suggests that users' interaction behavior can be explained and predicted through computational models.
PROCEEDINGS OF THE 2022 CHI CONFERENCE ON HUMAN FACTORS IN COMPUTING SYSTEMS (CHI' 22)
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Utkarsh Kunwar, Sheetal Borar, Moritz Berghofer, Julia Kylmala, Ilhan Aslan, Luis A. Leiva, Antti Oulasvirta
Summary: This paper discusses the challenges and problems of gesture recognition on smartwatches. By collecting diverse gesture data and utilizing prior knowledge, it is possible to design recognisers with robust performance on resource-constrained hardware.
IUI'22: 27TH INTERNATIONAL CONFERENCE ON INTELLIGENT USER INTERFACES
(2022)
