Autonomous Decision Research Articles

Enhancing multi-UAV air combat decision making via hierarchical reinforcement learning

In the realm of air combat, autonomous decision-making in regard to Unmanned Aerial Vehicle (UAV) has emerged as a critical force. However, prevailing autonomous decision-making algorithms in this domain predominantly rely on rule-based methods, proving challenging to design and implement optimal solutions in complex multi-UAV combat environments. This paper proposes a novel approach to multi-UAV air combat decision-making utilizing hierarchical reinforcement learning. First, a hierarchical decision-making network is designed based on tactical action types to streamline the complexity of the maneuver decision-making space. Second, the high-quality combat experience gained from training is decomposed, with the aim of augmenting the quantity of valuable experiences and alleviating the intricacies of strategy learning. Finally, the performance of the algorithm is validated using the advanced UAV simulation platform JSBSim. Through comparisons with various baseline algorithms, our experiments demonstrate the superior performance of the proposed method in both even and disadvantaged air combat environments.

Research on Lane-Changing Decision Making and Planning of Autonomous Vehicles Based on GCN and Multi-Segment Polynomial Curve Optimization.

This paper considers the interactive effects between the ego vehicle and other vehicles in a dynamic driving environment and proposes an autonomous vehicle lane-changing behavior decision-making and trajectory planning method based on graph convolutional networks (GCNs) and multi-segment polynomial curve optimization. Firstly, hierarchical modeling is applied to the dynamic driving environment, aggregating the dynamic interaction information of driving scenes in the form of graph-structured data. Graph convolutional neural networks are employed to process interaction information and generate ego vehicle's driving behavior decision commands. Subsequently, collision-free drivable areas are constructed based on the dynamic driving scene information. An optimization-based multi-segment polynomial curve trajectory planning method is employed to solve the optimization model, obtaining collision-free motion trajectories satisfying dynamic constraints and efficiently completing the lane-changing behavior of the vehicle. Finally, simulation and on-road vehicle experiments are conducted for the proposed method. The experimental results demonstrate that the proposed method outperforms traditional decision-making and planning methods, exhibiting good robustness, real-time performance, and strong scenario generalization capabilities.

Managing type 1 diabetes of a child: Parents' perspectives.

Managing type 1 diabetes (T1D) challenges children and their parents. Parents need to learn the necessary skills and later transfer the responsibility of care to their children as they develop. The transition process involves autonomy in behaviour and decision-making. This study explores the shared management experiences of Taiwanese parents and their children with type 1 diabetes. This study employed a qualitative design using a grounded theory approach. Purposive sampling was used at a medical centre in Taiwan for participant recruitment. Twenty-nine parents of children who had been diagnosed with T1D were interviewed in-depth. Data were analysed using constant comparison and repeated verification. After a child was diagnosed with T1D, the parents initiated 'Life-long lesson: Growing together with the child on the road to normality'. Three main categories emerged: 'confronting the disease diagnosis', 'establishing supportive and collaborative involvement' and 'assisting the child in building a sense of belonging'. Sub-categories within each significant category were also included. Taiwanese parents perhaps have a controlling or directive role for a long period in their child's lives and shared management of their health condition. This study's findings can help healthcare workers better understand the process of parents' shared management of T1D with their children and how to best communicate with children about the disease and care in accordance with the child's stage of development.

Post-clustering Prioritization Framework for Autonomous Decision Making in the Absence of Ground Truth via Hypothetical Probing

A generic prioritization framework is introduced for addressing the problem of automated prioritization of region of interest or target selection. The framework is based on the assumption that clustering of preliminary data for preidentified regions or targets of interest within an operational area has already occurred, i.e., post-classification, and that the clustering quality can be expressed as an energy/objective function. Region or target of interest prioritization then means to rank regions or targets of interest according to their probability of changing the energy/objective function value upon subsequent hypothetical probing as opposed to actually conducted reexamination, i.e., thorough follow-up or in-situ measurements. The mathematical formalism for calculating these probabilities to contribute to this change of the energy/objective function value is introduced and validated through numerical simulations. Moreover, these probabilities can also be understood as a confidence-check of the classification, i.e., the pre-clustering of the preliminary data. The operation of the prioritization framework is independent of the algorithm used to pre-cluster the preliminary data, and supports autonomous decision-making. It is widely applicable across many scientific disciplines and areas, ranging from the microscopic to the macroscopic scale. Due to its ability to help maximize scientific return while optimizing resource utilization, it is particularly relevant in the context of resource-constrained autonomous robotic planetary exploration as it may extend the Remaining Useful Lifetime (RUL) – a key aspect in Prognostics and Health Management (PHM) – of space missions. On a more general, PHM-relevant level, the prioritization framework may provide an additional mechanism of identifying and correcting the maintenance status of system components to assist predictive maintenance or condition-based maintenance.

Blockchain for the digital twin-driven autonomous optical network

In recent years, optical networks have become increasingly large-scale and complex, resulting in mounting pressures on operation and maintenance management. The techniques of the artificial intelligence (AI)-driven digital twin (DT) have been applied to optical networks to pave the way for creating the self-optimizing and self-controlling autonomous optical network (AON). However, the AON is confronted with many security and reliability issues: the centralized architecture is prone to data loss during system disconnection and active/standby switchover, and the massive data, DT models, and control instructions in the AON all face the risk of being tampered with, directly affecting the effectiveness of network regulation. In addition, the reliability of complex autonomous processes should be ensured. To address these challenges, in this paper, we applied blockchain to a DT-driven AON with the designed architecture and deployment scheme to realize data loss prevention, anti-tamper storage, secure autonomous operations, access control, and network element controller decisions. A blockchain-based solution was studied by considering the quality of transmission (QoT) estimation and link optimization as examples. Four chaincodes were designed and developed to perform the model storage, service definition, telemetry subscription, data collection, model implementation, and instruction delivery in the AON. Furthermore, the proposed solution was developed and deployed using Hyperledger Fabric. The Hyperledger Caliper was used to test the performance of the proposed blockchain system. The distributed environment is used for performance testing to simulate the real scenario. The test results are analyzed to indicate the future development direction.

Continuous Deep Sedation Until Death of Children at the End of Life: French Physicians' Opinions.

Objectives: To evaluate physicians' opinions concerning continuous deep sedation until death (CDSUD) and implementation of Claeys-Leonetti; a law intended to be applicable to all patients, but without a specific framework for children thus giving rise to ethically and legally complex situations. The secondary objective was to identify if physicians' characteristics could influence their opinions. Study Design: This was a national, multicenter, noninterventional cross-sectional survey from January 30, 2020, until March 1, 2020. The target population consisted of French physicians involved in children's end-of-life situations. The validated questionnaire explored respondents' characteristics and their opinions on four hypothetical pediatric clinical cases. Results: Analysis was conducted on 391 respondents. The oncological situation was more easily recognized as end of life compared with the neurological pathology (77% vs. 40.4%). Dependence on mechanical ventilation was another major factor influencing physicians in identifying end-of-life situations. Physicians clearly recognized the difference in intention between CDSUD and euthanasia. They accepted to implement CDSUD more easily in newborns. The withdrawal of artificial nutrition and hydration gave rise to divergent opinions. Respondents were in favor of adolescents' decision-making autonomy and their access to drafting advance directives. The child's best interest prevailed in case of objection by parents, except in situations outside the law's framework or in cases of disagreement within the health care team. Conclusion: Results of our study showed differences in the interpretation of the law concerning the CDSUD application framework and provide elements for reflection, which may ultimately contribute to the development of specific guidelines in CDSUD in children at the end of life.

Exploring challenges and perceptions in the learning environment: an online qualitative study of medical students

BackgroundThe teaching and learning environment (TLE) in medical schools is critical for shaping the outcomes and competencies of graduates. Research on TLE has highlighted its influence on student learning approaches and outcomes, yet gaps remain, particularly in qualitative insights, especially in Saudi Arabian contexts. This study aims to explore the students’ experiences and perceptions of the TLE in a new medical college.MethodsThis qualitative study consisted of a total of five focus group discussions (3consequtive sessions for male group and two for female group) conducted virtually using the Zoom videoconferencing application. All the discussion sessions took place during a lockdown because of the COVID-19 pandemic between December 2020 and February 2021. Each session lasted for 45–60 min. Each group was formed of 4–5 students from different academic levels in the Majmaah medical school, which was established 10 years ago and is located in a small city. After “verbatim transcription” of the sessions was made, a framework thematic analysis of the data was performed using the NVivo software.ResultsThe study results revealed that various elements, such as educational content, faculty roles, and personal factors, collectively influenced the students’ educational experiences. The students valued educational relevance and autonomous decision-making. The multifunctional role of faculty as mentors, evaluators, and resource providers was considered essential in enhancing academic experiences. Additionally, there was an evident need to improve the physical learning environment and facilities to adapt to emerging educational needs. These results align with existing literature, emphasizing the integration of theory and practice and the significant impact of faculty roles in academic experiences.ConclusionThe findings suggest that medical colleges should involve students more in decision-making related to their education and ensure the practical relevance of the educational content. Establishing open communication channels between students and faculty who act as mentors and evaluators is also essential. Furthermore, enhancing supportive infrastructures, such as mental health and financial services, and promoting extracurricular activities are crucial for fostering a more effective and nurturing learning environment.

Care considerations in medical assistance in dying for persons with mental illness as the sole underlying medical condition: a qualitative study of patient and family perspectives

BackgroundPersons with mental illness as their sole underlying medical condition are eligible to access medical assistance in dying (MAiD) in a small number of countries, including Belgium, the Netherlands, Luxemburg and Switzerland. In Canada, it is anticipated that people experiencing mental illness as their sole underlying medical condition (MI-SUMC) will be eligible to request MAiD as of March 17th 2024. To date, few studies have addressed patient and family perspectives on MAiD MI-SUMC care processes. This study aimed to address this gap and qualitatively explore the perspectives of persons with lived experience of mental illness and family members on care considerations during MI-SUMC implementation.MethodsThirty adults with lived experience of mental illness and 25 adult family members residing in Ontario participated in this study. To facilitate participant engagement, the semi-structured interview used a persona-scenario exercise to discuss perspectives on MAiD MI-SUMC acceptability and care considerations. Framework analysis was used to inductively analyze data using NVivo 12 Pro. Steps, processes, or other care considerations suggested by the participants were charted in a framework matrix after familiarization with the narratives. Key themes were further identified. A lived-experience advisory group participated in every aspect of this study.ResultsSix themes were developed from the patient and family narratives: (1) Raising MAiD MI-SUMC awareness; (2) Sensitive Introduction of MAiD MI-SUMC in goals of care discussions; (3) Asking for MAiD MI-SUMC: a person-focused response; (4) A comprehensive circle of MAiD MI-SUMC care; (5) A holistic, person-centered assessment process; and (6) Need for support in the aftermath of the decision. These themes highlighted a congruence of views between patient and family members and described key desired process ingredients, including a person-centred non-judgmental stance by care providers, inter-professional holistic care, shared decision making, and the primacy of patient autonomy in healthcare decision making.ConclusionsFamily and patient perspectives on the implementation of MAiD MI-SUMC offer important considerations for service planning that could complement existing and emerging professional practice standards. These stakeholders’ perspectives will continue to be essential in MAiD MI-SUMC implementation efforts, to better address the needs of diverse communities and inform improvement efforts.

A novel small object detection algorithm for UAVs based on YOLOv5

Due to the advances in deep learning, artificial intelligence is widely utilized in numerous areas. Technologies frontier, including computer vision, represented by object detection, have endowed unmanned aerial vehicles (UAVs) with autonomous perception, analysis, and decision-making capabilities. UAVs extensively used in numerous fields including photography, industry and agriculture, surveillance, disaster relief, and play an important role in real life. However, current object detection algorithms encountered challenges when it came to detecting small objects in images captured by UAVs. The small size of the objects, with high density, low resolution, and few features make it difficult for the algorithms to achieve high detection accuracy and are prone to miss and false detections especially when detecting small objects. For the case of enhancing the performance of UAV detection on small objects, a novel small object detection algorithm for UAVs adaptation based on YOLOv5s (UA-YOLOv5s) was proposed. (1) To achieve effective small-sized objects detection, a more accurate small object detection (MASOD) structure was adopted. (2) To boost the detection accuracy and generalization ability of the model, a multi-scale feature fusion (MSF) approach was proposed, which fused the feature information of the shallow layers of the backbone and the neck. (3) To enhance the model stability properties and feature extraction capability, a more efficient and stable convolution residual Squeeze-and-Excitation (CRS)module was introduced. Compared with the YOLOv5s, mAP@0.5 was achieved an impressive improvement of 7.2%. Compared with the YOLOv5l, mAP@0.5 increased by 1.0%, and GFLOPs decreased by 69.1%. Compared to the YOLOv3, mAP@0.5 decreased by 0.2% and GFLOPs by 78.5%. The study’s findings demonstrated that the proposed UA-YOLOv5s significantly enhanced the object detection performance of UAVs campared to the traditional algorithms.

The role of self-efficacy in women’s autonomy for health and nutrition decision-making in rural Bangladesh

BackgroundAgency – including the sub-domains of intrinsic agency, instrumental agency, and collective agency – is a critical component of the women’s empowerment process. Self-efficacy (a component of intrinsic agency) may operate as a motivational influence for women to make choices according to their own preferences or goals, such that higher self-efficacy would be associated with more autonomous decision-making (a key component of instrumental agency).MethodsWe examine these relationships using mixed methods. We developed a series of decision-making autonomy indices, which captured alignment between the woman’s reported and preferred roles in health and nutrition decisions. Using ordinal logistic regression, we assessed the relationship between generalized self-efficacy and decision-making autonomy.ResultsThere was a consistently positive association across all categories of decision-making, controlling for a number of individual and household-level covariates. In a sub-sample of joint decision-makers (i.e., women who reported making decisions with at least one other household member), we compared the association between generalized self-efficacy (i.e., one’s overall belief in their ability to succeed) and decision-making autonomy to that of domain-specific self-efficacy (i.e., one’s belief in their ability to achieve a specific goal) and decision-making autonomy. Across all decision-making categories, domain-specific self-efficacy was more strongly associated with decision-making autonomy than generalized self-efficacy. In-depth interviews provided additional context for interpretation of the regression analyses.ConclusionsThe results indicate the importance of the role of self-efficacy in the women’s empowerment process, even in the traditionally female-controlled areas of health and nutrition decision-making. The development of the decision-making autonomy index is an important contribution to the literature in that it directly recognizes and captures the role of women’s preferences regarding participation in decision-making.

Abstract WP72: National Assessment of Trainee and Non-Vascular Neurology Acute Stroke Decision-Making at StrokeNet Regional Coordinating Centers

Background: The impact of acute stroke treatment changes, including development of Vascular Neurology, advances in neuroimaging, and endovascular treatment (EVT), on stroke team trainee utilization and decision-making are unknown. For insight into high-performance teams, we surveyed StrokeNet Regional Coordinating Centers (RCCs). Methods: A confidential, self-administered, pilot-tested survey was distributed to the site PIs of all 27 StrokeNet RCCs. The survey assessed in-person acute stroke response staffing, involvement of trainees, decisional autonomy, and attitudes regarding decision-makers. Absent data fields excluded. Descriptive statistics presented. Results: 18 of 27 RCCs completed the survey (gross response rate 66%). 67% (12/18) reported trainees (Resident and/or Fellow) always respond to ED and inpatient activations and 39% (7/18) that trainees always respond to EVT activations. 44% (7/16) and 27% (4/15) reported Nurse Practioner (NP) and Physician Assistant (PA) involvement in ED stroke responses, respectively. Thrombolytic decision-making authority (prior to attending staffing) rested with Fellows and Residents in 61% (11/18) and 18% (3/17), respectively. No NP or PA independent use was identified (0%). 89% (16/18) reported attending Emergency Physicians did not order thrombolytics prior to stroke team consultation. Apart from Vascular Neurology, which was universally acceptable, acceptability by attending specialty for thrombolytic ordering was: Emer Med (88%), Gen Neurology (75%), Neurosurgery (33%), Fam Med (12%), Int Med (11%) and Radiology (6%). Acceptability for EVT decisions was Neurosurgery (94%), Gen Neurology (59%), Radiology (50%), Emer Med (44%), Fam Med (12%) and Int Med (11%). Conclusions: At StrokeNet RCCs, acute stroke teams have significant trainee involvement in both ED and inpatient acute stroke responses and allow trainees significant autonomy. Stroke team involvement likely occurs early in the ED, as few reported attending emergency physicians ordering thrombolytics prior to consultation. Respondents varied in acceptability of different specialties as decision-makers for both thrombolytics and EVT. These findings have potential workforce, education, and training implications.

“It Means Being Safe, Having Power over One's Body”: Black Female Adolescent Experiences with Reproductive Health Counseling in Primary Care

Study ObjectiveThere is limited research examining the sexual and reproductive health (SRH) care needs and experiences of Black adolescents who are assigned female at birth (AFAB). This study aimed to understand the perspectives of Black AFAB adolescents in their receipt of SRH counseling in primary care and elicit preferences for SRH-related communication with clinicians. MethodsWe interviewed English-speaking Black AFAB adolescents, ages 13-17, living in North Carolina between February and June 2022 about their SRH care experiences. The interviews were conducted via video conferencing, audio-recorded, professionally transcribed, and analyzed using a thematic approach. ResultsWe interviewed 23 adolescents (mean age 15.8 years) across 10 geographically diverse counties in North Carolina. Most conveyed positive perceptions of clinicians regarding trust and comfort. However, many expressed concerns about clinicians not doing enough to ensure patient confidentiality, provide them with enough information to make informed and autonomous decisions, and destigmatize discussing SRH issues during the clinical encounter. Suggested improvements include sociodemographic congruence between patients and clinicians (eg, younger Black women), ensuring time alone with clinicians across adolescence, and proactively sharing information to promote respect and autonomy. ConclusionAlthough Black AFAB adolescents had positive perceptions of their health care providers regarding relational components of care, many participants shared significant gaps and areas for improvement in the quality of care received to better align services with their preferences for SRH care. These findings can inform the development of SRH counseling best practices and clinician education to support respect and autonomy, which are routinely denied to Black AFAB adolescents.

Designing policies that could work: understanding the interaction between policy design spaces and organizational responses in public sector

The goal of this paper is to contribute toward bridging the gap between policy design and implementation by focusing on domains, such as education, healthcare and community services, where policy implementation is largely left to the autonomous decision of public service providers, which are strategic actors themselves. More specifically, we suggest that two characteristics of policy design spaces in which policies are designed, i.e., the level of ideational coherence and the prevailing function of the adopted policy instruments, generate systematic patterns of responses in terms of the extent of compliance with policy goals, the presence of strategic gaming and possible defiance. We illustrate our model through a contrastive case study of the introduction of performance-based funding in the higher education sector in four European countries (France, Italy, Norway, and the United Kingdom). Our analysis displays that policy designs chosen by governments to steer public systems have different trade-offs in terms of responses of the public organizations involved that are essential to effectively implement governmental policies. The model we are proposing provides therefore a framework to understand how these interactions unfold in specific contexts, what are their effects on the achievement of policy goals and how policymakers could exploit their degrees of freedom in policy design to reduce unwanted effects.

Adaptive AUV Mission Control System Tested in the Waters of Baffin Bay

The primary objectives of this paper are to test an adaptive sampling method for an autonomous underwater vehicle, specifically tailored to track a hydrocarbon plume in the water column. An overview of the simulation of the developed applications within the autonomous system is presented together with the subsequent validation achieved through field trials in an area of natural oil seeps near to Scott Inlet in Baffin Bay. This builds upon our prior published work in methodological development. The method employed involves an integrated backseat drive of the AUV, which processes in situ sensor data in real time, assesses mission status, and determines the next task. The core of the developed system comprises three modular components—Search, Survey, and Sample—each designed for independent and sequential execution. Results from tests in Baffin Bay demonstrate that the backseat drive operating system successfully accomplished mission goals, recovering water samples at depths of 20 m, 50 m, and 200 m before mission completion and vehicle retrieval. The principal conclusion drawn from these trials underscores the system’s resilience in enhanced decision autonomy and validates its applicability to marine pollutant assessment and mitigation.

IntelliHome: The Automated Household Infrastructure

Abstract: The rapid advancement of technology in recent years has ushered in an era of unprecedented connectivity and automation in our daily lives. One of the most promising developments in this domain is the concept of IntelliHome – an integrated and intelligent household infrastructure that seamlessly combines smart devices, artificial intelligence, and data analytics to enhance convenience, efficiency, and sustainability within the home environment. This research paper delves into the multifaceted realm of IntelliHome, exploring its technological underpinnings, potential benefits, challenges, and societal implications. The paper begins by elucidating the fundamental components of an IntelliHome, encompassing a spectrum of smart devices such as thermostats, lighting systems, security systems, and appliances, all interconnected through the Internet of Things (IoT). It discusses the pivotal role of artificial intelligence in orchestrating these devices, enabling autonomous decision-making, predictive analytics, and adaptive customization to cater to the unique needs and preferences of each household. Furthermore, the research paper delves into the manifold benefits of an IntelliHome, including energy conservation, increased security, enhanced convenience, and improved quality of life. It examines how IntelliHome systems can reduce carbon footprints, optimize resource utilization, and contribute to the development of sustainable smart cities. Finally, the research paper contemplates the broader societal implications of IntelliHome technology. It explores how these systems may redefine the boundaries of work, leisure, and domesticity, and how they may influence family dynamics, social interactions, and community engagement. It also investigates the economic aspects and economic impacts.

Research of microprocessor device and software for remote control of a robotic system

The modern stage of the development of intelligent robotic systems is characterized by the expansion of fields of application, which is due to autonomous work and decision-making in conditions of uncertainty. The object of research is the system of remote control of robotic systems. During the remote control of robotic systems, problems arise that are associated with the use of wireless communication in real time. The article analyzes software and hardware implementations of various remote control systems suitable for use as part of autonomous robotic systems and analyzes promising microcontroller platforms for implementing a remote control device for a robotic system. A brief review of existing protocols for transmitting control signals using radio communication equipment and microprocessor platforms for the development of embedded systems is performed, among which a solution is selected for research. Several approaches to the control of a robotic system are highlighted – control using a wired connection and corresponding protocols, control via wireless communication or via the Internet, control via general-purpose network protocols. The target platform is chosen and justified, and the S.BUS protocol is analyzed with the provision of an algorithm for obtaining the values of the control channels from the S.BUS package. The structure and algorithm of functioning of the microprocessor remote control system based on the ESP32 microcontroller and the FreeRTOS OS are given. A study of the operation process of the proposed remote control system is carried out, for which it is placed on the chassis of a ground autonomous robotic system with four-wheel drive, and the delay time of the control signal from the receiver to the engine control modules is determined. According to the conducted analysis, the expediency of using specialized radio communication equipment with the S.BUS protocol for controlling executive devices as part of a robotic system, for precise movement control in real time, is shown.

Artificial Intelligence Gamers Based on Deep Reinforcement Learning

This study investigates the design and implementation of Artificial Intelligence (AI) game players based on deep reinforcement learning, offering a novel approach to autonomous decision-making and strategy acquisition in intelligent games. Initially, the fundamental principles and algorithms of deep reinforcement learning are introduced, along with the fusion of deep learning and reinforcement learning. Subsequently, existing research is reviewed, and the pros and cons of current methodologies are examined, highlighting the underlying issues and challenges. The utilization of AI players in mainstream games is then introduced, and the influence of AI players on contemporary games is analyzed. Through this analysis of AI players in mainstream games, the strengths and weaknesses of current AI players are identified, and recommendations for optimizing them are provided. This study holds significant implications for guiding the design and development of intelligent game players, while also enriching the application of deep reinforcement learning within the gaming domain.

Qualitative study on domestic social robot adoption and associated security concerns among older adults in Slovenia

IntroductionDespite the increasing use of domestic social robots by older adults, there remains a significant knowledge gap regarding attitudes, concerns, and potential adoption behavior in this population. This study aims to categorize older adults into distinct technology adoption groups based on their attitudes toward domestic social robots and their behavior in using the existing technology.MethodsAn exploratory qualitative research design was used, involving semi-structured interviews with 24 retired Slovenian older adults aged 65 years or older, conducted between 26 June and 14 September 2023.ResultsFour distinct groups of older adults were identified: (1) Cautious Optimists, (2) Skeptical Traditionalists, (3) Positive Optimists, and (4) Technophiles based on eight characteristics.DiscussionThese groups can be aligned with the categories of the Diffusion of Innovation Theory. Privacy and security concerns, influenced by varying levels of familiarity with the technology, pose barriers to adoption. Perceived utility and ease of use vary considerably between groups, highlighting the importance of taking into account the different older adults. The role of social influence in the adoption process is complex, with some groups being more receptive to external opinions, while others exhibit more autonomous decision-making.

Framework for Autonomous UAV Navigation and Target Detection in Global-Navigation-Satellite-System-Denied and Visually Degraded Environments

Autonomous Unmanned Aerial Vehicles (UAVs) have possible applications in wildlife monitoring, disaster monitoring, and emergency Search and Rescue (SAR). Autonomous capabilities such as waypoint flight modes and obstacle avoidance, as well as their ability to survey large areas, make UAVs the prime choice for these critical applications. However, autonomous UAVs usually rely on the Global Navigation Satellite System (GNSS) for navigation and normal visibility conditions to obtain observations and data on their surrounding environment. These two parameters are often lacking due to the challenging conditions in which these critical applications can take place, limiting the range of utilisation of autonomous UAVs. This paper presents a framework enabling a UAV to autonomously navigate and detect targets in GNSS-denied and visually degraded environments. The navigation and target detection problem is formulated as an autonomous Sequential Decision Problem (SDP) with uncertainty caused by the lack of the GNSS and low visibility. The SDP is modelled as a Partially Observable Markov Decision Process (POMDP) and tested using the Adaptive Belief Tree (ABT) algorithm. The framework is tested in simulations and real life using a navigation task based on a classic SAR operation in a cluttered indoor environment with different visibility conditions. The framework is composed of a small UAV with a weight of 5 kg, a thermal camera used for target detection, and an onboard computer running all the computationally intensive tasks. The results of this study show the robustness of the proposed framework to autonomously explore and detect targets using thermal imagery under different visibility conditions. Devising UAVs that are capable of navigating in challenging environments with degraded visibility can encourage authorities and public institutions to consider the use of autonomous remote platforms to locate stranded people in disaster scenarios.

Artificial intelligence vs. autonomous decision-making in streaming platforms: A mixed-method approach

Although the empowerment of technology is of great value to society, little is known about its downstream effects on consumers' decisions. This research draws on the expectation–confirmation theory and autonomy in artificial intelligence (AI) and investigates how AI (vs. autonomous choice) has detrimental effects on consumer outcomes, creating an autonomy-technology tension — i.e., the conflict arising from AI technology diminishing consumers' autonomy in their choices. Four studies using a mixed-method approach reveal that the use of AI recommendations in streaming platforms creates an autonomy-technology tension that reduces consumers' performance expectancy, thus lowering their satisfaction. However, such effects are contingent on the nature of the AI recommendations. While a mismatch between AI recommendations and consumer preferences might backfire, AI's negative effect is mitigated when choices match consumers' preferences. We make significant theoretical and practical contributions to empirical research on consumers' sense of autonomy while interacting with AI.