High-fidelity Simulation Environment Research Articles

Hovering of Bi-Directional Motor Driven Flapping Wing Micro Aerial Vehicle Based on Deep Reinforcement Learning

Inspired by hummingbirds and certain insects, flapping wing micro aerial vehicles (FWMAVs) exhibit potential energy efficiency and maneuverability advantages. Among them, the bi-directional motor-driven tailless FWMAV with simple structure prevails in research, but it requires active pose control for hovering. In this paper, we employ deep reinforcement learning to train a low-level hovering strategy that directly maps the drone’s state to motor voltage outputs. To our knowledge, other FWMAVs in both reality and simulations still rely on classical proportional-derivative controllers for pose control. Our learning-based approach enhances strategy robustness through domain randomization, eliminating the need for manually fine-tuning gain parameters. The effectiveness of the strategy is validated in a high-fidelity simulation environment, showing that for an FWMAV with a wingspan of approximately 200 mm, the center of mass is maintained within a 20 mm radius during hovering. Furthermore, the strategy is utilized to demonstrate point-to-point flight, trajectory tracking, and controlled flight of multiple drones.

Dynamics and advanced active disturbance rejection control of tethered UAV

Due to its vast potential for applications in agricultural harvesting, the tethered unmanned aerial vehicle (UAV) has garnered considerable interest. However, the anti-disturbance control of this system remains fraught with obstacles, stemming primarily from its inherent nonlinearity, unpredictable airflow disturbances, unmeasurable tether tension, and the changes in the center of gravity during operation. We propose a comprehensive dynamic model for the operational tethered unmanned aerial vehicle with considering the center of gravity shift. Because of obstructions caused by branches and leaves, the tethered UAV can't reach fruits inside the tree. Therefore, two deformation functions of the tethered quadcopter are considered. Additionally, in order to enhance the effectiveness of traditional Active Disturbance Rejection Control, we present an optimized controller combining a phase compensation algorithm with a nonlinear function. The proposed Controller is validated in a high-fidelity simulation environment, considering internal interference, wind field interference and reactive force from produces. Besides, the simulation is specifically developed for picking operation, The results underscore the better observation capabilities of the enhanced controller proposed in this study. Moreover, when deployed in a simulated picking environment, the improved controller exhibits remarkable resistance to the interference and significantly mitigates chattering issues.

A practical real-time control strategy for high-performance building HVAC operation concerning potential pandemic outbreaks in post-pandemic era

The lessons learned from the COVID-19 pandemic emphasize the need for building HVAC systems to be adaptable and prepared to adjust their operation for infection risk mitigation for future pandemics. Most guidelines suggest increasing outdoor air fraction in HVAC systems during pandemics. However, this approach is not energy-efficient due to the conditioning of additional outdoor air, and many practical systems face limitations in introducing more outdoor air than normal due to capacity or design constraints. Therefore, there is an urgent need for an alternative solution for HVAC system operation during pandemics that is energy-efficient, can effectively mitigate infection risks and does not require extensive retrofits to existing systems. To meet this need, this study proposes a practical real-time control strategy for high-performance HVAC operation in response to potential pandemic outbreaks. First, a supply air temperature (SAT) reset scheme for infection risk mitigation is proposed based on Mamdani fuzzy inference theory. Then, a coordinated control scheme is proposed to coordinate SAT reset with chilled water temperature adjustment for energy-efficiency enhancement. This strategy does not require system equipment retrofits but only changes in control logic to transition to pandemic operation. The effectiveness of the proposed strategy is validated in a high-fidelity Modelica-based simulation environment under various weather conditions. Results show that, compared to normal operation, the proposed strategy can significantly reduce new infection cases by 32∼46 % without increasing energy use, and may even achieve slight energy savings of 9∼13 %, while maintaining expected occupant comfort.

Investigation of Lithium-Ion Battery Negative Pulsed Charging Strategy Using Non-Dominated Sorting Genetic Algorithm II

To address the critical issue of polarization during lithium-ion battery charging and its adverse impact on battery capacity and lifespan, this research employs a comprehensive strategy that considers the charging duration, efficiency, and temperature increase. Central to this approach is the proposal of a novel negative pulsed charging technique optimized using the Non-Dominated Sorting Genetic Algorithm II (NSGA-II). This study initiates the creation of an intricate electrothermal coupling model, which simulates variations in internal battery parameters throughout the charging cycle. Subsequently, NSGA-II is implemented in MATLAB to fine-tune pulsed charging and discharging profiles, generating a Pareto front showcasing an array of optimal solutions tailored to a spectrum of goals. Leveraging the capabilities of the COMSOL Multiphysics software 6.2 platform, a high-fidelity simulation environment for lithium-ion battery charging is established that incorporates three charging strategies: constant-current (CC) charging, a multi-stage constant-current (MS-CC) charging protocol, and a pulsed-current (PC) charging strategy. This setup works as a powerful instrument for assessing the individual effects of these strategies on battery characteristics. The simulation results strongly support the superiority of the proposed pulsed-current charging strategy, which excels in increasing the battery temperature and amplifying battery charge capacity. This dual achievement not only bolsters charging efficiency significantly but also underscores the strategy’s potential to augment both the practical utility and long-term viability of lithium-ion batteries, thereby contributing to the advancement of sustainable energy storage solutions.

Estimating Uncertainties from Dual-Doppler Radar Measurements of Onshore Wind Plants Using LES

Towards the ongoing work of improving the capability of flow modeling within and around wind plants, an onshore model validation benchmark campaign is underway based on a field experiment involving multiple wind plants in Oklahoma, U.S.A. Dual-Doppler radar is being leveraged to provide flowfield information for the benchmarking owing to the unparalleled capability of such radar to capture minute-by-minute horizontal wind fields over a scale of tens of kilometers. However, dual-Doppler radar exhibits sampling artifacts that must be considered during model validation, and these are due to probe-volume averaging, coarse azimuthal/elevational resolution, non-ideal stereo angles, and coarse temporal sampling. Such sources of error in radar-reconstructed flowfields can be quantified using virtual radar sampling in the high-fidelity simulation environment (i.e., large-eddy simulation (LES)) where the true velocity field is known with confidence, and this is the uncertainty quantification approach adopted in this article. We leverage a virtual radar tool designed to replicate the specific sampling strategy of the X-band dual-Doppler instrument installed in the field campaign. This tool is featured in LES of an expansive 100 km by 100 km region of Oklahoma including hundreds of wind turbines modeled as actuator disks. In agreement with the sampling principles of radar, the results show that large-scale flow structures are qualitatively well-resolved by the instrument, though more simulation time and analysis are needed to determine the accuracy of the radar’s integral lengthscale estimates. At the turbine scale, the radar struggles to capture all of the features of the turbine wakes. The process of probe-volume averaging, as well as the subsequent interpolation to a Cartesian grid, biases the reconstruction of the peak near-wake deficit by an average of 2 m/s, or around 20% of the freestream velocity. Errors in this quantity of interest, as well as one characterizing the magnitude of the free-flow wind speed around the wakes, are found to be sensitive to the radar beam-crossing angle and beam range.

Enhanced Traffic Light Guidance for Safe and Energy-Efficient Driving: A Study on Multiple Traffic Light Advisor (MTLA) and 5G Integration

This paper presents Multiple Traffic Light Advisor (MTLA), a novel Green Light Optimal Speed Advisory (GLOSA) system that leverages 5G communication technology. GLOSA systems are emerging as a key component in intelligent transportation systems, thanks to the development of effective communication technologies. At its core, MTLA serves as a guidance system for drivers, providing real-time instructions to adjust vehicle speed to optimize the utilization of current and future states of traffic lights along their route.The work addresses several limitations in the current state-of-the-art approaches, including the use of an overly simplified velocity profile, the omission of potential grip and jerk in problem formulation, and the absence of a detailed description of the algorithm’s implementation aspects. Initially, we comprehensively present an optimization-free implementation of the overall control architecture based on an unconventional speed profile. Subsequently, MTLA is improved within a non-linear Model Predictive Control (MPC) framework which uses the latter nonoptimal solution as an initial guess and considers potential grip and jerk in the problem formulation. The developed systems are numerically tested and compared within a high-fidelity simulation environment using the IPG CarMaker simulator. The results demonstrate promising performance in terms of energy savings, with a significant reduction of 37% in energy usage, as well as improved overall comfort with respect to the case where no guidance is given to the driver. These findings suggest a high potential for future developments in this domain.

A Real-to-Sim-to-Real Approach for Vision-Based Autonomous MAV-Catching-MAV

This paper studies the task of vision-based MAV-catching-MAV, where a catcher MAV (micro aerial vehicle) can detect, localize, and pursue a target MAV autonomously. Since it is challenging to develop detectors that can effectively detect unseen MAVs in complex environments, the main novelty of this paper is to propose a real-to-sim-to-real approach to address this challenge. In this method, images of real-world environments are first collected. Then, these images are used to construct a high-fidelity simulation environment, based on which a deep-learning detector is trained. The merit of this approach is that it allows efficient automatic collection of large-scale and high-quality labeled datasets. More importantly, since the simulation environment is constructed from real-world images, this approach can effectively bridge the sim-to-real gap, enabling efficient deployment in real environments. Another contribution of this paper lies in the successful implementation of a fully autonomous vision-based MAV-catching-MAV system including proposed estimation and pursuit control algorithms. While the previous works mainly focused on certain aspects of this system, we developed a completely autonomous system that integrates detection, estimation, and control algorithms on real-world robotic platforms.

An enhanced motorcycle tyre model characterised through experimental riding data

Tyre-road interaction governs motorcycle dynamics; however, the most widespread tyre model formulations must be characterised through a dedicated test bench on the lab or road, unavailable to many interested subjects. This article proposed a new tyre model formulation, conceived to be characterised through riding data using standard instrumentation. Albeit its coefficients are identified through quasi-static, uncombined slip manoeuvres, the model addresses transient, combined manoeuvres and is adaptive to road friction levels and static weight through statistical relationships from the literature. A pre-existing formulation was improved and expanded. The model’s behaviour in demanding conditions was investigated through a high-fidelity simulation environment, using a Magic Formula tyre model as the reference. Next, the characterisation procedure was carried out using actual riding data. The model’s accuracy is shown by reproducing numerically one of the manoeuvres and through comparison with the results of a bench test. The proposed model could correctly reproduce the primary behaviour of a Magic Formula model, also concerning tyre moments and steering torque. Characterising the tyre model through real riding data proved feasible, and its robust formulation limited the propagation of estimation errors. The proposed tyre model formulation and characterisation procedure should interest, among others, those subjects that lack access to a tyre testing machine.

Advances in remote and cloud-based simulation through web conferencing platforms.

Simulation-based learning has become an essential element in entry-level perfusion education. While the use of simulation has been demonstrated to improve patient outcomes, few institutions possess the budgetary resources to build and maintain a high-fidelity simulation environment. This project aims to identify novel uses of web conferencing platforms to support in-person, remote, and virtual simulation exercises. The Zoom Virtual Meeting platform (Zoom Video Communications, Inc.) was incorporated into the perfusion simulation curriculum at the Medical University of South Carolina CVP Program. Among the observed benefits of incorporating meeting platform software included high-definition audio/visual outputs and recording capabilities, 3-D remote simulation, remote simulation proctoring, and classroom-based learning. Additional Zoom features included remote controlled screen access and sharing, annotations, and break-out rooms and activities. The combination of high-fidelity simulation with virtual and remote features may enhance the learning experience in healthcare education. Future developments in technology and software, simulation education, and instructions through virtual/remote learning may provide a pathway for the future of cardiovascular perfusion education.

Reactive optimal motion planning to anywhere in the presence of moving obstacles

In this paper, a novel optimal motion planning framework that enables navigating optimally from any initial, to any final position within confined workspaces with convex, moving obstacles is presented. Our method outputs a smooth velocity vector field, which is then employed as a reference controller in order to sub-optimally avoid moving obstacles. The proposed approach leverages and extends desirable properties of reactive methods in order to provide a provably convergent and safe solution. Our algorithm is evaluated with both static and moving obstacles in synthetic environments and is compared against a variety of existing methods. The efficacy and applicability of the proposed scheme is finally validated in a high-fidelity simulation environment.

Predictive Path-Tracking Control of an Autonomous Electric Vehicle with Various Multi-Actuation Topologies.

This paper presents the development of path-tracking control strategies for an over-actuated autonomous electric vehicle. The vehicle platform is equipped with four-wheel steering (4WS) as well as torque vectoring (TV) capabilities, which enable the control of vehicle dynamics to be enhanced. A nonlinear model predictive controller is proposed taking into account the nonlinearities in vehicle dynamics at the limits of handling as well as the crucial actuator constraints. Controllers with different actuation formulations are presented and compared to study the path-tracking performance of the vehicle with different levels of actuation. The controllers are implemented in a high-fidelity simulation environment considering scenarios of vehicle handling limits. According to the simulation results, the vehicle achieves the best overall path-tracking performance with combined 4WS and TV, which illustrates that the over-actuation topology can enhance the path-tracking performance during conditions under the limits of handling. In addition, the performance of the over-actuation controller is further assessed with different sampling times as well as prediction horizons in order to investigate the effect of such parameters on the control performance, and its capability for real-time execution. In the end, the over-actuation control strategy is implemented on a target machine for real-time validation. The control formulation proposed in this paper is proven to be compatible with different levels of actuation, and it is also demonstrated in this work that it is possible to include the particular over-actuation formulation and specific nonlinear vehicle dynamics in real-time operation, with the sampling time and prediction time providing a compromise between path-tracking performance and computational time.

Towards ubiquitous and nonintrusive measurements of brain function in the real world: assessing blink-related oscillations during simulated flight using portable low-cost EEG.

Blink-related oscillations (BRO) are newly discovered neurophysiological phenomena associated with spontaneous blinking and represent cascading neural mechanisms including visual sensory, episodic memory, and information processing responses. These phenomena have been shown to be present at rest and during tasks and are modulated by cognitive load, creating the possibility for brain function assessments that can be integrated seamlessly into real-world settings. Prior works have largely examined the BRO phenomenon within controlled laboratory environments using magnetoencephalography and high-density electroencephalography (EEG) that are ill-suited for real-world deployment. Investigating BROs using low-density EEG within complex environments reflective of the real-world would further our understanding of how BRO responses can be utilized in real-world settings. We evaluated whether the BRO response could be captured in a high-fidelity flight simulation environment using a portable, low-density wireless EEG system. The effects of age and task demands on BRO responses were also examined. EEG data from 30 licensed pilots (age 43.37 +/- 17.86, 2 females) were collected during simulated flights at two cognitive workload levels. Comparisons of signal amplitudes were undertaken to confirm the presence of BRO responses and mixed model ANOVAs quantified the effects of workload and age group on BRO amplitudes. Significant increases in neural activity were observed post-blink compared to the baseline period (p < 0.05), confirming the presence of BRO responses. In line with prior studies, results showed BRO time-domain responses from the delta band (0.5-4 Hz) consisting of an early negative peak followed by a positive peak post-blink in temporal and parietal electrodes. Additionally, task workload and age-related effects were also found, with observations of the enhancement of BRO amplitudes with older age and attenuation of BRO responses in high workloads (p < 0.05). These findings demonstrate that it is possible to capture BRO responses within simulated flight environments using portable, low-cost, easy-to-use EEG systems. Furthermore, biological and task salience were reflected in these BRO responses. The successful detection and demonstration of both task-and age-related modulation of BRO responses in this study open the possibility of assessing human brain function across the lifespan with BRO responses in complex and realistic environments.

Deep learning of vehicle dynamics

Recent advances in deep-learning-based identification of dynamic systems have resulted in a new generation of approaches utilizing state-space neural models with innovation noise structure, improved reformulation of multiple shooting, batch optimization, and a subspace identification-inspired form of encoders. The latter is used to learn a reconstructability map to estimate model states from past inputs and outputs. By using the SUBNET approach, which belongs to the state-of-the-art of these methods, we show how to effectively use these approaches to identify reliable vehicle models from data both in continuous and discrete time, respectively. We showcase the approach on the identification of the dynamics of a Crazyflie 2.1 nano-quadcopter and an F1tenth electric car both in a high-fidelity simulation environment, and in case of the electric car, on real measured data. The results indicate that new-generation of deep-learning methods offer Efficient system identification of vehicle dynamics in practice.

LSTM-Based Virtual Load Sensor for Heavy-Duty Vehicles.

In this paper, a special recurrent neural network (RNN) called Long Short-Term Memory (LSTM) is used to design a virtual load sensor that estimates the mass of heavy vehicles. The estimation algorithm consists of a two-layer LSTM network. The network estimates vehicle mass based on vehicle speed, longitudinal acceleration, engine speed, engine torque, and accelerator pedal position. The network is trained and tested with a data set collected in a high-fidelity simulation environment called Truckmaker. The training data are generated in acceleration maneuvers across a range of speeds, while the test data are obtained by simulating the vehicle in the Worldwide harmonized Light vehicles Test Cycle (WLTC). Preliminary results show that, with the proposed approach, heavy-vehicle mass can be estimated as accurately as commercial load sensors across a range of load mass as wide as four tons.

Surgical and Radiology Trainees' Proficiency in Reading Mammograms: the Importance of Education for Cancer Localisation.

Medical imaging with mammography plays a very important role in screening and diagnosis of breast cancer, Australia's most common female cancer. The visualisation of cancers on mammograms often forms a diagnosis and guidance for radiologists and breast surgeons, and education platforms that provide real cases in a simulated testing environment have been shown to improve observer performance for radiologists. This study reports on the performance of surgical and radiology trainees in locating breast cancers. An enriched test set of 20 mammography cases (6 cancer and 14 cancer free) was created, and 18 surgical trainees and 32 radiology trainees reviewed the cases via the Breast Screen Reader Assessment Strategy (BREAST) platform and marked any lesions identifiable. Further analysis of performance with high- and low-density cases was undertaken, and standard metrics including sensitivity and specificity. Radiology trainees performed significantly better than surgical trainees in terms of specificity (0.72 vs. 0.35; P < 0.01). No significant differences were observed between the surgical and radiology trainees in sensitivity or lesion sensitivity. Mixed results were obtained with participants regarding breast density, with higher density cases generally having lower performance. The higher specificity of the radiology trainees compared to the surgical trainees likely represents less exposure to negative mammography cases. The use of high-fidelity simulated self-test environments like BREAST is able to benchmark, understand and build strategies for improving cancer education in a safe environment, including identifying challenging scenarios like breast density for enhanced training.

Observational Study of Conformity in Yet Another Medical Learning Environment: Conformity to Preceptors During High-Fidelity Simulation.

Altering one's behavior to comply with inaccurate suggestions made by others (i.e., conformity) has been studied since the 1950s. Although several studies have documented its occurrence in medical education, it has yet to be examined in a high-fidelity simulation environment. It was hypothesized that a large majority of learners would conform to a preceptor. A total of 42 student dyads (a medical student paired with a resident) participated in one of four clinical scenarios to manage the diagnosis and treatment of a simulated patient encounter. Once the learners became familiar with the patient's case, a preceptor entered the simulation, offered an equivocal suggestion about diagnosis or management, and then left. Two raters observed the video recordings of how the learners managed the case after this suggestion was made. The nature of these interactions was also documented. Sixteen (38.10%) of the 42 medical student dyads conformed to the equivocal information presented by the preceptors. Observations of these interactions showed that all of the medical students conformed to the residents, but not all of the medical students conformed to the preceptors. Many learners conform to preceptors by acting on their equivocal suggestion when managing a patient case during high-fidelity simulation.

Hierarchical Reinforcement Learning for Air Combat at DARPA's AlphaDogfight Trials

Autonomous control in high-dimensional, continuous state spaces is a persistent and important challenge in the fields of robotics and artificial intelligence. Because of high risk and complexity, the adoption of AI for autonomous combat systems has been a long-standing difficulty. In order to address these issues, DARPA's AlphaDogfight Trials (ADT) program sought to vet the feasibility of and increase trust in AI for autonomously piloting an F-16 in simulated air-to-air combat. Our submission to ADT solves the high-dimensional, continuous control problem using a novel hierarchical deep reinforcement learning approach consisting of a high-level policy selector and a set of separately trained low-level policies specialized for excelling in specific regions of the state space. Both levels of the hierarchy are trained using off-policy, maximum entropy methods with expert knowledge integrated through reward shaping. Our approach outperformed human expert pilots and achieved a second-place rank in the ADT championship event. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Impact Statement–</i> Significant performance milestones in reinforcement learning have been achieved in recent years, with autonomous agents demonstrating super-human performance across a wide variety of tasks. Before these algorithms can be extensively deployed in real-world defense applications, a greater level of trust must first be achieved. ADT was an important step towards developing the trust necessary to operationalize these algorithms, by demonstrating their effectiveness on a foundational yet relevant problem in a high-fidelity simulation environment. Developed for the program, our hierarchical reinforcement learning agent was designed alongside of and competed against active fighter pilots, and ultimately defeated a graduate of the United States Air Force's F-16 Weapons Instructor Course in match play.

Can We Build it? Yes, We Can! Development Procedure of High-Fidelity Simulation Environments for Human-Agent Teams

This paper presents a bottom-up approach to designing and developing a high-fidelity simulation environment that fosters human acceptance of artificial intelligence (AI) as teammates by refining their mental models of appropriate teamwork expectations. Our process begins by first identifying an appropriate contextual situation that warrants humans teaming with AI as opposed to other team-based configurations. Those criteria are based on the delegation of roles appropriate to the established strengths of humans/AI, as well as the interdependence created between those roles to accentuate the expectations of teammate behavior. Next, qualitative interviews should be conducted with diverse subject matter experts to gain a comprehensive understanding of the situation and the environmental attributes through various perspectives. Once analyzed using thematic analysis, themes present themselves as design recommendations on how to create a high-fidelity simulation environment that nurtures human-agent team collaboration.

Sequencing Patient Care Tasks as an Infection Prevention Practice

Healthcare-associated infections are common, yet largely preventable. We examined whether nurses that successfully limited their contamination spread in a high-fidelity simulated environment sequenced their tasks differently than nurses that spread more contamination by reanalyzing an existing dataset. In the simulations, contamination spread was tracked using live viral surrogates (bacteriophages), which are harmless to humans. An overall contamination performance score was calculated for each participant, who were divided into one of three performance groups: high (M = 93%), medium (M = 78%), or low (M = 59%). An ANOVA showed contamination performance group did not have a statistically significant effect on the order nurses completed tasks; the largest effect size was small ( ηG2 = 0.019). Thus, even if nurses that successfully limit their contamination spread do sequence their tasks differently, it may not be practically meaningful because it is a small effect.

Effect of Mixed Reality on Delivery of Emergency Medical Care in a Simulated Environment

Mixed-reality (MR) technology has the potential to enhance care delivery, but there remains a paucity of evidence for its efficacy and feasibility. To assess the efficacy and feasibility of MR technology to enhance emergency care delivery in a simulated environment. This pilot randomized crossover trial was conducted from September to November 2021 at a single center in a high-fidelity simulated environment with participants block randomized to standard care (SC) or MR-supported care (MR-SC) groups. Participants were 22 resident-grade physicians working in acute medical and surgical specialties prospectively recruited from a single UK Academic Health Sciences Centre. Data were analyzed from September to December 2022. Participants resuscitated a simulated patient who was acutely unwell, including undertaking invasive procedures. Participants completed 2 scenarios and were randomly assigned to SC or MR-SC for the first scenario prior to crossover. The HoloLens 2 MR device provided interactive holographic content and bidirectional audiovisual communication with senior physicians in the MR-SC group. The primary outcome was error rate assessed via the Imperial College Error Capture (ICECAP) multidimensional error-capture tool. Secondary outcomes included teamwork (Observational Teamwork Assessment for Surgery [OTAS]; range, 0-6 and Teamwork Skills Assessment for Ward Care [T-SAW-C]; range, 1-5), scenario completion, stress and cognitive load (NASA Task Load Index [NASA-TLX; range 0-100]), and MR device user acceptability. A total of 22 physicians (15 males [68.2%]; median [range] age, 28 [25-34] years) were recruited. MR technology significantly reduced the mean (SD) number of errors per scenario compared with SC (5.16 [3.34] vs 8.30 [3.09] errors; P = .003), with substantial reductions in procedural (0.79 [0.75] vs 1.52 [1.20] errors; P = .02), technical (1.95 [1.40] vs 3.65 [2.03] errors; P = .01), and safety (0.37 [0.96] vs 0.96 [0.85] errors; P = .04) domains. MR resulted in significantly greater scenario completion rates vs SC (22 scenarios [100%] vs 14 scenarios [63.6%]; P = .003). It also led to significant improvements in the overall quality of teamwork and interactions vs SC as measured by mean (SD) OTAS (25.41 [6.30] vs 16.33 [5.49]; P < .001) and T-SAW-C (27.35 [6.89] vs 18.37 [6.09]; P < .001) scores. As reported via mean (range) NASA-TLX score, there were significant reductions for MR-SC vs SC in participant temporal demands (38 [20-50] vs 46 [30-70]; P = .03) and significant improvements in self-reported task performance (50 [30-60] vs 39 [10-70]; P = .01). Overall, 19 participants (86.4%) reported that they were more confident in making clinical decisions and undertaking clinical procedures with MR support. This study found that the use of MR technology reduced error, improved teamwork, and enhanced practitioner confidence when used to support the delivery of simulated emergency medical care. ClinicalTrials.gov Identifier: NCT05870137.