Modality Switching Research Articles

Salient Object Detection From Arbitrary Modalities.

Toward desirable saliency prediction, the types and numbers of inputs for a salient object detection (SOD) algorithm may dynamically change in many real-life applications. However, existing SOD algorithms are mainly designed or trained for one particular type of inputs, failing to be generalized to other types of inputs. Consequentially, more types of SOD algorithms need to be prepared in advance for handling different types of inputs, raising huge hardware and research costs. Differently, in this paper, we propose a new type of SOD task, termed Arbitrary Modality SOD (AM SOD). The most prominent characteristics of AM SOD are that the modality types and modality numbers will be arbitrary or dynamically changed. The former means that the inputs to the AM SOD algorithm may be arbitrary modalities such as RGB, depths, or even any combination of them. While, the latter indicates that the inputs may have arbitrary modality numbers as the input type is changed, e.g. single-modality RGB image, dual-modality RGB-Depth (RGB-D) images or triple-modality RGB-Depth-Thermal (RGB-D-T) images. Accordingly, a preliminary solution to the above challenges, i.e. a modality switch network (MSN), is proposed in this paper. In particular, a modality switch feature extractor (MSFE) is first designed to extract discriminative features from each modality effectively by introducing some modality indicators, which will generate some weights for modality switching. Subsequently, a dynamic fusion module (DFM) is proposed to adaptively fuse features from a variable number of modalities based on a novel Transformer structure. Finally, a new dataset, named AM-XD, is constructed to facilitate research on AM SOD. Extensive experiments demonstrate that our AM SOD method can effectively cope with changes in the type and number of input modalities for robust salient object detection. Our code and AM-XD dataset will be released on https://github.com/nexiakele/AMSODFirst.

Examining modal and amodal language processing in proficient bilinguals: evidence from the modality-switch paradigm.

Recent discussions have emphasized the significance of embodied processing in language comprehension. Nevertheless, continuous debates persist regarding the relative contribution of modal (embodied) and amodal (abstract) processing of language. The current study investigated the contribution of modal processing in the first (L1) and second (L2) language, hypothesizing higher level of abstract amodal symbol processing in L2 than L1, since the correspondence of L1 and L2 (i.e., the symbol-to-symbol assignment) is in the foreground when learning L2. We employed the modality-switch paradigm (Pecher et al., 2003) in both German and French versions with proficient sequential German and French bilinguals (N = 79). Participants were presented with noun-adjective pairs (e.g., keys - jingling) in both languages and decided whether the adjective could be applied to the noun. This task repeatedly requires switching modality between trials, (e.g., from auditory [keys - jingling] to olfactory [soap - perfumed]), typically causing switch costs on response latency as compared to maintaining the modality. Contrary to the hypothesis, we observed modality switch effects (MSE) in both L1 and L2. This result suggests that embodied language processing occurs not only in the first language but also extends to the second language thus challenging the assumption that L2 processing predominantly involves abstract amodal symbol processing. Notably, however, significant L1 and L2 MSEs were found for French, whereas for German already the L1 effect was rather weak (though significant); the corresponding L2 effect was not significant. Thus, the results hinted at differences between languages regarding the relative role of modal and amodal processing.

Social presence and collaborative creativity in leaner media

As people increasingly rely on virtual spaces for both personal and professional exchanges, the dynamics of how we connect, share ideas, and work together have evolved. Online interactions often engage in modality switching where communication starts in one modality, such as talking, then switches to another, such as texting. We examined how modality switching affected creativity and social presence when collaborating using computer mediated communication. Participants first met via an audio or text modality, then switched to the other modality. They then engaged in a collaborative story creation task. Using the Social Information Processing Theory for our theoretical framework, we found that switching from text to audio benefitted creativity. We found no differences between texted and audio communication in social presence, desire for future interaction, or creativity. Replicating others’ results, we did find differences in conversational balance: Texted conversations were more balanced than audio conversations. We newly discovered that modality switching affected balance. Audio conversations were more balanced when they followed text interactions rather than preceded them. Ultimately, more balanced conversations – regardless of modality – led to greater creativity in dyads.

Age-related Electrophysical Correlates of Cross-modal Attention Switching.

The human experience demands seamless attentional switches between sensory modalities. Aging raises questions about how declines in auditory and visual processing affect cross-modal attention switching. This study used a cued cross-modal attention-switching paradigm where visual and auditory stimuli were simultaneously presented on either spatially congruent or incongruent sides. A modality cue indicated the target modality, requiring a spatially left versus right key-press response. EEG recordings were collected during task performance. We investigated whether the mixing costs (decreased performance for repetition trials in a mixed task compared with a single task) and switch costs (decreased performance for a switch of target modality compared with a repetition) in cross-modal attention-switching paradigms would exhibit similarities in terms of behavioral performance and the ERP components to those observed in the traditional unimodal attention-switching paradigms. Specifically, we focused on the ERP components: cue-locked P3 (mixing/switch-related increased positivity), target-locked P3 (mixing/switch-related decreased positivity), and target-locked lateralized readiness potential (mixing/switch-related longer latency). In addition, we assessed how aging impacts cross-modal attention-switching performance. Results revealed that older adults exhibited more pronounced mixing and switch costs than younger adults, especially when visual and auditory stimuli were presented on incongruent sides. ERP findings showed increased cue-locked P3 amplitude, prolonged cue-locked P3 latency, decreased target-locked P3 amplitude, prolonged target-locked P3 latency in association with switch costs, and prolonged onset latency of the target-locked lateralized readiness potential in association with the mixing costs. Age-related effects were significant only for cue-locked P3 amplitude, cue-locked P3 latency (switch-related), and target-locked P3 latency (switch-related). These findings suggest that the larger mixing costs and switch costs in older adults were because of the inefficient use of modality cues to update a representation of the relevant task sets and required more processing time for evaluating and categorizing the target.

Mo-dalities, Mo Problems: Impacts of Provider Modality Switching in Hybrid Outpatient Clinics

Mo-dalities, Mo Problems: Impacts of Provider Modality Switching in Hybrid Outpatient Clinics

Should we meet in person or on Facetime?

The present study drew from expectancy violations theory, predicted outcome value (POV) theory, and the modality switching perspective to examine online daters who subsequently met for FtF-like communication. Hierarchical regression analyses tested whether online daters’ post-modality switch (MS) expectedness ratings, evaluations, and POV forecasts were related to their: (a) pre-MS length of online association; (b) number of pre-MS partner photos seen; (c) use of pre-MS phone calls; and (d) decision to hold the first FtF-like meeting in person or through video chat. Behavioral and physical appearance expectedness ratings were positively related to the number of pre-MS partner photos seen, and curvilinearly (inverted-u shape) related to daters’ pre-MS length of association. Physical appearance evaluations were positively associated with the number of pre-MS photos seen, while behavioral and physical appearance evaluations displayed positive associations with pre-MS phone calls. Finally, post-MS POV forecasts were positively associated with the number of pre-MS partner photos seen, holding the MS through video chat, post-MS assessments of a partner’s behavioral expectedness, and post-MS evaluations of a partner’s behavioral and physical appearance. An interaction also emerged for behavioral expectedness and evaluation on POV, which implied that negative expectancy violations were more useful than positive violations in predicting a relationship’s potential viability.

Romantic relationship initiation and escalation through mobile dating apps: Affordances, modality weaving, and paradoxical beliefs

Like online dating sites, mobile dating applications are popular technologies for navigating the dating market, whether for seeking romantic relationships or sexual partners. The searching-matching-interacting (SMI) framework describes mate selection in the dating market and how mediated market intermediaries (e.g., dating apps) can aid these functions. We conducted in-depth interviews ( N = 37) to explore dating app use during relationship initiation and relationship escalation, examining the SMI framework. For searching and matching, perceptions of affordances such as visibility, searchability, and locatability governed participants’ selection and use of apps. Examining the interacting function, our findings provide some of the first empirical support for modality weaving during relationship development. Rather than a simple modality switch from app to face-to-face, participants reported integrating multiple interpersonal and masspersonal channels, including social media, texting, and video. Channel transitions were made based on goals (e.g., relationship escalation, verification, uncertainty reduction) and perceived affordances (e.g., synchronicity, editability, bandwidth, accessibility). Notably, participants’ evaluations of the advantages and disadvantages of dating apps revealed several tensions and paradoxes among beliefs and behaviors. These paradoxes indicate why many users may not satisfy interpersonal goals such as hookups or long-term romantic relationships through mobile dating apps. Our findings support the viability of the SMI framework, and we extend its theorizing for studying relationship initiation and relationship development.

Impact of aging on crossmodal attention switching

Previous studies on crossmodal visual-auditory attention switching using a spatial discrimination task showed performance costs when the target modality changed relative to when it repeated. The present study (n = 42 for each age group) examined age-related changes in crossmodal attention switching by asking young (age range 19 to 30 years old) and older (age range 64 to 80 years old) participants to respond to unimodal central cues and bimodal lateralized stimuli. The participants’ task was to indicate the location of the target in the relevant modality using button presses. Results showed general attention switch costs. Additionally, we found no specific age-related increase of attention switch costs (no difference in performance between switch and repetition of target modality), but age-related increased mixing costs (decreased performance for repetition in modality-mixed condition compared to single target modality). Moreover, spatial distraction produced a crossmodal congruency effect, which was only slightly larger in older adults. Taken together, age-related increased mixing costs suggest a general difficulty with maintaining more than one task, but no specific age-related crossmodal impairment in crossmodal attention switching.

Construction of wear-resistant visible-infrared-compatible camouflage film and its spectral control mechanisms

The ceaseless advancement of modern surveillance technologies has underscored an urgent need for the development of stealth materials with multi-spectral camouflage capabilities. Spectral selectivity, achieved through the design of micro-nano structures, can effectively manage conflicting requirements across various light regimes. Nonetheless, creating visible-Infrared (IR)-compatible devices that are both structurally simple and suitable for use as optic windows or coatings remains a formidable challenge. Here, we report a MgF2/Ag/MgF2 visible-IR-compatible film based on Monte Carlo optimization which has high visible transparency, low IR emissivity, ideal wear-resistance, and dynamic tunability. The designed film can maintain high transparency (∼80%) to ensure replication of background without sacrificing low IR emissivity (∼0.3). Notably, the film maintains its spectral selectivity across a wide range of angles and is endowed with the capability for dynamic modality switching between “heating” and “cooling” states under electrical actuation with good stability under long-term service and harsh environment. This film has a maximum temperature characteristic modulation range of 3.5 °C, demonstrating excellent performance in intelligent thermal management. It is also compatible with electrochromic camouflage devices enabling it to cover a broader spectrum of reconnaissance without contradictions. It is an effective way to fulfill the camouflage requirements and smart thermal management across diverse environments, and possesses great potential for scalability and flexibility.

Unveiling the drivers of modal switch from motorcycles to public transport in Southeast Asia

Most studies on commuter and public transport mode choice rely on theoretical frameworks that draw boundaries between the utility function, social-ecological system, demographic and socio-economic attributes, and psychological theory. While models predicting the shift towards public transport are commonly applied in developed countries, there is a lack of similar studies that construct these models in developing countries. In addition, in some developing countries, the motorcycle is one of the main private motorised travel modes. Therefore, the modal shift from motorcycles to public transport in Southeast Asia to reduce private mobility is one of the important topics to examine. This paper presents a systematic literature review, utilising a comprehensive search strategy across electronic databases and reputable sources to identify directed acyclic graphs (DAGs), revealing factors influencing the choice and intention to shift to public transport. Data synthesis from selected studies highlights the intrinsic and extrinsic variables influencing public transport use, motorcycle use, and the modal shift to public transport. The study also proposes a theoretical framework for the modal shift in usage from motorcycles to public transport.

Multi-channel broadband nonvolatile programmable modal switch.

Mode-division multiplexing (MDM) in chip-scale photonics is paramount to sustain data capacity growth and reduce power consumption. However, its scalability hinges on developing efficient and dynamic modal switches. Existing active modal switches suffer from substantial static power consumption, large footprints, and narrow bandwidth. Here, we present, for the first time, to the best of our knowledge, a novel multiport, broadband, non-volatile, and programmable modal switch designed for on-chip MDM systems. Our design leverages the unique properties of integrating nanoscale phase-change materials (PCM) within a silicon photonic architecture. This enables independent manipulation of spatial modes, allowing for dynamic, non-volatile, and selective routing to six distinct output ports. Crucially, our switch outperforms current dynamic modal switches by offering non-volatile, energy-efficient multiport functionality and excels in performance metrics. Our switch exhibits exceptional broadband operating bandwidth exceeding 70 nm, with low loss (< 1 dB), and a high extinction ratio (> 10 dB). Our framework provides a step forward in chip-scale MDM, paving the way for future green and scalable data centers and high-performance computers.

Efficient bicycle networks and expansion strategies

&#x0D; &#x0D; &#x0D; This article explains some of the key attributes and factors relating to efficient bicycle networks and their expansion strategies. Varying from a network scale to brief discussion about the streetscape design, many scientific publications are referenced to explain the crucial aspects which must be taken into consideration when planning an attractive cycling network. Modal switch from other transport modes, especially from cars is one of the goals of this cycling promotion. Various methods such as building new bicycle facilities, improving the quality of existing ones and increasing the comfort and safety of the connections were found to improve the attractiveness of cycling in urban context. Additionally, this article analyses the reasons behind cyclists’ route choice as well as the metrics such as connectedness, directedness and coverage which can be used to measure the level of network functionality. Lastly, network growth under limited resources is considered and an equitable distribution of cycleways across the city is found to be important in preventing neighbourhood segregation while not being too far from a utilitarian kind of distribution in terms of return to investment. &#x0D; &#x0D; &#x0D;

Cross-Modal Object Tracking via Modality-Aware Fusion Network and a Large-Scale Dataset.

Visual object tracking often faces challenges such as invalid targets and decreased performance in low-light conditions when relying solely on RGB image sequences. While incorporating additional modalities like depth and infrared data has proven effective, existing multimodal imaging platforms are complex and lack real-world applicability. In contrast, near-infrared (NIR) imaging, commonly used in surveillance cameras, can switch between RGB and NIR based on light intensity. However, tracking objects across these heterogeneous modalities poses significant challenges, particularly due to the absence of modality switch signals during tracking. To address these challenges, we propose an adaptive cross-modal object tracking algorithm called modality-aware fusion network (MAFNet). MAFNet efficiently integrates information from both RGB and NIR modalities using an adaptive weighting mechanism, effectively bridging the appearance gap and enabling a modality-aware target representation. It consists of two key components: an adaptive weighting module and a modality-specific representation module. The adaptive weighting module predicts fusion weights to dynamically adjust the contribution of each modality, while the modality-specific representation module captures discriminative features specific to RGB and NIR modalities. MAFNet offers great flexibility as it can effortlessly integrate into diverse tracking frameworks. With its simplicity, effectiveness, and efficiency, MAFNet outperforms state-of-the-art methods in cross-modal object tracking. To validate the effectiveness of our algorithm and overcome the scarcity of data in this field, we introduce CMOTB, a comprehensive and extensive benchmark dataset for cross-modal object tracking. CMOTB consists of 61 categories and 1000 video sequences, comprising a total of over 799K frames. We believe that our proposed method and dataset offer a strong foundation for advancing cross-modal object-tracking research. The dataset, toolkit, experimental data, and source code will be publicly available at: https://github.com/mmic-lcl/ Datasets-and-benchmark-code.

Environmental impacts of bicycling in urban areas: A micro-simulation approach

We develop a microsimulation model of urban transport for a city of a medium size and use it to evaluate the impacts of modal switch to bicycle. An activity-based approach is used to generate daily transportation schedules for several groups of households. We first consider the case of mixed traffic, where the bicycles and cars share the same lanes, and find that a significant modal switch to bicycle has indisputable benefits on road congestion and emissions of pollutant gases. We then consider the development of cycle-paths, where the two modes run on separate lanes, and find that it improves the benefits obtained under mixed traffic. At the same time, our analysis shows that a small modal switch (below 15%) to bicycle does not necessarily produce the expected benefits. This is because in uncongested road sections, the bicycles cause delays to other vehicles.

Sequential adaptation to modality incompatibility.

Modality-compatible stimulus-response mappings (e.g., responding vocally to an auditory stimulus and manually to a visual stimulus) are often easier to perform than modality-incompatible sets (reversed modality mappings). Here, we investigate sequential, trial-to-trial, modulations of modality compatibility effects. By reanalyzing a previous experiment and conducting two specifically tailored, new experiments, we demonstrate robust within-task sequential modulations. Furthermore, we test for between-task adaptations by intermixing the modality switching task with a Simon task. Results show reliable sequential adaptations within the modality switching task, but no transfer of adaptation between tasks in either direction. We discuss how a combination of prominent theoretical accounts such as conflict adaptation and episodic binding can serve as the cognitive underpinnings of the observed sequential adaptations. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

#4784 IDENTIFICATION OF FACTORS ASSOCIATED WITH DEATH IN DIALYSIS PATIENTS USING A MACHINE LEARNING-BASED PREDICTIVE MODEL

Abstract Background and Aims Few studies including unplanned dialysis starts have used machine learning for the prediction of death in dialysis patients. The objective of the study was to use R software algorithms to develop machine learning predictive models for the identification of death-related factors in patients undergoing hemodialysis (HD) and peritoneal dialysis (PD). Method This study included adult patients undergoing HD and PD started in a planned or urgent manner in a dialysis center between January 2014 and January 2019. Epidemiological, clinical and laboratory data were collected. Univariate analysis was followed by ML-based analyses. Then, multivariate regressions were obtained using stepwise and Cox regression analyses. Finally, a Random Forest predictive model was generated after variables with missing values &amp;gt;30% were removed. Results Of 581 patients included, 170 died (29,2%). On univariate analysis death was associated with age, number of comorbidities, dialysis modality switching, creatinine, PTH and albumin values at dialysis initiation, presence of diabetes (DM), hospitalization, function recovery and central venous catheter (CVC) for dialysis access. Patients who started dialysis with a CVC had a worse survival (p = 0.0034) than those who did not use CVC, started HD with AVF, or received PD (Figure 1). Data were split into 20% for testing the regression model, and 80% for training the model. Data preprocessing for Cox regression included imputing some values using bag impute (decision trees), creating dummy variables, and removing collinear variables. Death was associated with older age (p &amp;lt; 0.001), fewer ESI-free months (p &amp;lt; 0.001) and lower initial creatinine (p = 0.008) (Table 1). The model C-index was 0.8099. Random forest ranked the following variables predictive of death in descending order of importance: ESI-free months; age; initial levels of creatinine, PTH and albumin; number of comorbidities; dialysis-related infection; initial phosphorus and hemoglobin; hospitalizations; male gender; modality switching (Figure 2). The agreement of the model obtained was 0.8110. Conclusion ESI-free months, age and initial levels of creatinine were associated with death on both multivariate and ML-based analyses.

#4731 THE CHANGING HAEMODIALYSIS POPULATION IN A SINGLE UK CENTRE: A POST PANDEMIC PHENOMENON?

Abstract Background and Aims Following the Covid-19 pandemic, UK dialysis centres have seen a sharp increase in numbers of patients requiring renal replacement therapy (RRT) particularly haemodialysis (HD). This has resulted in pressures on dialysis capacity and staffing. There may also have been negative effects on other aspects of quality of care, such as a fall in the proportion of patients dialysing via an Arterio-venous fistula (AVF) or worsening anaemia. To study this further, we conducted a retrospective observational study to describe the incident and prevalent HD populations in the post-Covid pandemic period and compare these against pre-pandemic patterns, in a single UK dialysis centre. Method We reviewed data on prevalent HD patient numbers across our service. We specifically collected data from all patients commencing HD between January 2021 and December 2022 using electronic medical records. We collected information on time known to renal prior to starting dialysis, clinic data including which types of clinic attended, documentation of pre-dialysis decisions about modality as well as vascular access referrals, vascular access clinics and vascular access surgery as well as haemoglobin (Hb) at first dialysis. We compared figures with published pre-pandemic data from the 23rd annual UK Renal Registry report on data collected to December 2019 for our centre. We defined a planned start to dialysis as patients who were known to renal &amp;gt;90 days. This aligns with the registry definition. We compared proportional data using the chi square test. Results Comparing prevalent data in Dec. 2022 with Dec 2019, there was a 17% increase in the prevalent HD population from 298 to 349. The proportion of patients dialysing via an AVF fell from 88% to 80% (p = 0.006). The spread across different types of HD in Dec.2022 remained static at 71% In-Centre HD (72% in Dec 2019), 11% Satellite HD (9% in 2019) and 17% HHD (18% 2019) with overall numbers rising across all HD modalities (p for trend = 0.37). PD numbers also rose from 58 to 71 keeping the total percentage on home therapies stable at 31% in our unit. From Jan. 2021- Dec. 2022 inclusive 147 people commenced HD of whom 115 (78%) had planned starts. This compares to 58 starting HD in 2019 and 79% planned (p = 0.85). There was a significant fall in the proportion of planned start HD patients who commenced dialysis with an AVF/AVG, from 81% in 2019 to 56% in 2021-22 (p = 0.003). Themes emerged n planned starters who did not start HD with an AVF/AVG and are detailed in Figure 1. Median Hb level of planned starts on HD was lower at only 94 for 2021-22 and 99 in 2019. Looking in detail at those missing appointments in clinics, 33% had prior documented evidence of fear of Covid-19. Delays in AVF surgery consisted of delayed and cancelled theatre dates, failed AVF requiring re-do or AVF awaiting further intervention at time of dialysis initiation. In the group with a last minute change in decision sighted reasons included: worsening eyesight, development of dementia in partner, reducing mobility. Others changed their mind after further input from the PD team. Despite attending the Advanced Kidney Clinic (AKC) there were still a number of patients with no decision at all. Cancellation of group sessions for dialysis education persisted in our centre until mid-2022 and may have contributed to reduced timely decision making as well as late modality switches. All those with an early failure of PD (defined as &amp;lt; 90 days) had chosen PD prior to the Covid-19 pandemic with intervening remote visits in AKC. Conclusion Following the Covid-19 pandemic, unexpected and unprecedented changes in the HD incident population have been observed, resulting in pressures on dialysis services. In particular, we report an increase in absolute number of dialysis patients, a significant fall in the proportion of patients initiating dialysis with definitive vascular access and a rise in those starting anaemic. This was not related to a change in the proportion of patients with planned starts but may be related to delayed shared decision making as a result of remote consultations and greater delays in vascular access surgery. We have identified a number of areas for service improvement.

In vivo co-registered hybrid-contrast imaging by successive photoacoustic tomography and magnetic resonance imaging

Magnetic resonance imaging (MRI) and photoacoustic tomography (PAT) offer two distinct image contrasts. To integrate these two modalities, we present a comprehensive hardware-software solution for the successive acquisition and co-registration of PAT and MRI images in in vivo animal studies. Based on commercial PAT and MRI scanners, our solution includes a 3D-printed dual-modality imaging bed, a 3-D spatial image co-registration algorithm with dual-modality markers, and a robust modality switching protocol for in vivo imaging studies. Using the proposed solution, we successfully demonstrated co-registered hybrid-contrast PAT-MRI imaging that simultaneously displays multi-scale anatomical, functional and molecular characteristics on healthy and cancerous living mice. Week-long longitudinal dual-modality imaging of tumor development reveals information on size, border, vascular pattern, blood oxygenation, and molecular probe metabolism of the tumor micro-environment at the same time. The proposed methodology holds promise for a wide range of pre-clinical research applications that benefit from the PAT-MRI dual-modality image contrast.

Hundred-watt level high-order mode all-fiberized random distributed feedback Raman fiber laser with high mode purity

An all-fiberized random distributed feedback Raman fiber laser (RRFL) with LP11 mode output at 1134 nm has been demonstrated experimentally, where an intracavity acoustically induced fiber grating is employed for modal switching. The maximum output power of LP11 mode is 93.8 W with the modal purity of 82%, calculated by numerical mode decomposition technology based on stochastic parallel-gradient descent algorithm. To our best knowledge, this is the highest output power with high purity of LP11 mode generated from the RRFL. This work may pave a path towards advanced fiber lasers with special temporal and spatial characteristics for applications.

Optimal Generator Policy for Hybrid Fuel UAV under Airspace Noise Restrictions

Here we study an optimal control problem involving energy management of a hybrid-fuel Unmanned Aerial Vehicle (UAV). The planning problem for a hybrid-fuel platform involves determining the path while managing the energy resources, which includes a policy for power modality switching whenever applicable. The hybrid-fuel platform considered here involves a generator and battery pack combined in a series fashion as energy sources on-board a UAV. Also included in the problem are the noise restrictions, which place constraints on generator operation depending on the airspace location. These emulate possible restrictions on UAV noise that occur in military surveillance missions or in urban path planning, where the collective noise of many UAVs, some with combustion engines, may be restricted in certain areas or times of the day. We present a hybrid methodology which starts from an initial path and generator pattern obtained from a mixed integer linear program (MILP) solution. The generator pattern from the discrete solution is then refined in an optimal control framework with an objective to minimize fuel usage, while considering the nonlinear battery and generator dynamics and noise-restriction constraints. Optimal control problem is solved with a nonlinear program solver, IPOPT. Numerical results are presented and analyzed with varying path lengths and scenarios. This work aims to serve as an initial study of this hybrid-fuel UAV problem within an optimal control framework, which can be extended to refinement of both the generator pattern and the trajectory in tandem, while considering vehicle and power dynamics that are often ignored in discrete path planning solutions.