Good Trajectories Research Articles

In silico validation of hyaluronic acid – drug conjugates based targeted drug delivery for the treatment of COVID-19

The impact of COVID-19 urges scientists to develop targeted drug delivery to manage Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) viral infections with a fast recovery rate. The aim of the study is to develop Hyaluronic Acid (HA) drug conjugates of viral drugs to target two important enzymes (Mpro and PLpro) of SARS-CoV-2. Three antiviral drugs, namely Dexamethasone (DEX), Favipiravir (FAV), and Remdesivir (REM), were chosen for HA conjugation due to their reactive functional groups. Free forms of drugs (DEX, FAV, REM) and HA drug conjugates (HA-DEX, HA-FAV, HA-REM, HA-RHA, HA-RHE) were validated against Mpro (PDB ID 6LU7) and PLpro (PDB 7LLZ), which play an essential role in the replication and reproduction of the SARS-CoV-2 virus. The results of the present study revealed that HA-drug conjugates possess higher binding affinity and the best docking score towards the Mpro and PLpro target proteins of SARS-CoV-2 than their free forms of drugs. ADMET screening resulted that HA-drug conjugates exhibited better pharmacokinetic profiles than their pure forms of drugs. Further, molecular dynamic simulation studies, essential dynamics and free energy landscape analyses show that HA antiviral drug conjugates possess good trajectories and energy status, with the PLpro target protein (PDB 7LLZ) of SARS-CoV-2 through long-distance (500 ns) simulation screening. The research work recorded the best drug candidate for Cell-Targeted Drug Delivery (CTDD) for SARS-CoV-2-infected cells through hyaluronic acid conjugates of antiviral drugs. Communicated by Ramaswamy H. Sarma

Trust and monetary policy

AbstractWe analyze how trust affects the transmission of negative demand and supply shocks using a behavioral macroeconomic model. We define trust to have two dimensions: trust in the central bank's inflation target and trust in the central bank's capacity to stabilize the business cycle. We find, first, that when large negative shocks occur, the subsequent trajectories taken by output gap and inflation typically coalesce around a good and a bad trajectory. Second, these good and bad trajectories are correlated with movements in trust. In the bad trajectories, trust collapses, and in the good trajectories, it is not affected. This feature is stronger when a negative supply shock occurs than in the case of a negative demand shock. Third, initial conditions, in particular the initial state of inflation and output expectations, matter. Unfavorable initial expectations drive the economy into a bad trajectory, and favorable initial expectations produce good trajectories. Fourth, we analyze the sensitivity of our results with respect to the size of the shocks. Fifth, we derive implications of our results for our capacity of making forecasts about the effects of large demand and supply shocks.

Learning from Good Trajectories in Offline Multi-Agent Reinforcement Learning

Offline multi-agent reinforcement learning (MARL) aims to learn effective multi-agent policies from pre-collected datasets, which is an important step toward the deployment of multi-agent systems in real-world applications. However, in practice, each individual behavior policy that generates multi-agent joint trajectories usually has a different level of how well it performs. e.g., an agent is a random policy while other agents are medium policies. In the cooperative game with global reward, one agent learned by existing offline MARL often inherits this random policy, jeopardizing the utility of the entire team. In this paper, we investigate offline MARL with explicit consideration on the diversity of agent-wise trajectories and propose a novel framework called Shared Individual Trajectories (SIT) to address this problem. Specifically, an attention-based reward decomposition network assigns the credit to each agent through a differentiable key-value memory mechanism in an offline manner. These decomposed credits are then used to reconstruct the joint offline datasets into prioritized experience replay with individual trajectories, thereafter agents can share their good trajectories and conservatively train their policies with a graph attention network (GAT) based critic. We evaluate our method in both discrete control (i.e., StarCraft II and multi-agent particle environment) and continuous control (i.e., multi-agent mujoco). The results indicate that our method achieves significantly better results in complex and mixed offline multi-agent datasets, especially when the difference of data quality between individual trajectories is large.

Paediatric end-of-life care at home.

A terminally ill child should have the possibility to be at home with their family during the end of life. Provision of care from primary care nurses (PCNs) is crucial, but no model exists on how specialised paediatric palliative care teams (SPPCTs) support the PCNs to perform this task. To investigate how PCNs evaluated a shared care model between a SPPCT and PCNs in paediatric end-of-life care. A 23-item questionnaire was distributed to PCNs involved in care of 14 terminally ill children in November 2019 and January 2020. Descriptive statistics were used. A total of 20 questionnaires were returned from nurses who agreed/completely agreed that an introductory meeting made them better prepared (78.9%) to deal with the death of a child in their care, cooperating with family members (70.6%) and dealing with own emotions (73.7%). A total of 69.2% felt the meeting helped coping with parents' pressure and 88.9% stated that the trajectory changed how they saw themselves involved in future paediatric palliative care. The shared care model was evaluated positively. Clear agreements and specialist support were prerequisites for good trajectories at the end of life. Further research is needed to investigate if the shared care model optimises palliative care and security in relation to child and family.

Surgical, survival and quality of life outcomes in over 1000 pelvic exenterations: lessons learned from a large Australian case series

To determine surgical, survival and quality of life outcomes across different tumour streams and lessons learned over 28 years. Consecutive patients undergoing pelvic exenteration at a single, high volume, referral hospital, between 1994 and 2022 were included. Patients were grouped according to their tumour type at presentation as follows, advanced primary rectal cancer, other advanced primary malignancy, locally recurrent rectal cancer, other locally recurrent malignancy and non-malignant indications. The main outcomes included, resection margins, postoperative morbidity, long-term overall survival, and quality of life outcomes. Non-parametric statistics and survival analyses were performed to compare outcomes between groups. Of the 1023 pelvic exenterations performed, 981 (95.9%) unique patients were included. Most patients underwent pelvic exenteration due to locally recurrent rectal cancer (N= 321, 32.7%) or advanced primary rectal cancer (N= 286, 29.2%). The rates of clear surgical margins (89.2%; P< 0.001) and 30-days mortality were higher in the advanced primary rectal cancer group (3.2%; P= 0.025). The 5-year overall survival rates were 66.3% in advanced primary rectal cancer and 44.6% in locally recurrent rectal cancer. Quality of life outcomes differed across groups at baseline, but generally had good trajectories thereafter. International benchmarking revelled excellent comparative outcomes. The results of this study demonstrate excellent outcomes overall, but significant differences in surgical, survival and quality of life outcomes across patients undergoing pelvic exenteration due to different tumour streams. The data reported in this manuscript can be utilized by other centres as benchmarking as well as proving both subjective and objective outcome details to support informed decision-making for patients.

Missile guidance with assisted deep reinforcement learning for head-on interception of maneuvering target

In missile guidance, pursuit performance is seriously degraded due to the uncertainty and randomness in target maneuverability, detection delay, and environmental noise. In many methods, accurately estimating the acceleration of the target or the time-to-go is needed to intercept the maneuvering target, which is hard in an environment with uncertainty. In this paper, we propose an assisted deep reinforcement learning (ARL) algorithm to optimize the neural network-based missile guidance controller for head-on interception. Based on the relative velocity, distance, and angle, ARL can control the missile to intercept the maneuvering target and achieve large terminal intercept angle. To reduce the influence of environmental uncertainty, ARL predicts the target’s acceleration as an auxiliary supervised task. The supervised learning task improves the ability of the agent to extract information from observations. To exploit the agent’s good trajectories, ARL presents the Gaussian self-imitation learning to make the mean of action distribution approach the agent’s good actions. Compared with vanilla self-imitation learning, Gaussian self-imitation learning improves the exploration in continuous control. Simulation results validate that ARL outperforms traditional methods and proximal policy optimization algorithm with higher hit rate and larger terminal intercept angle in the simulation environment with noise, delay, and maneuverable target.

An Adaptive Control for the Path Tracking of an Active Leg Prosthesis

This paper presents an adaptive fuzzy-PID control strategy applied to an active lower limb prosthesis for trajectories tracking in normal walking, stairs climbing, and stairs descent. Trying to imitate a natural human limb, the prosthesis design challenges rehabilitating amputees to resume normal activities. A dynamic model of an ankle-knee active prosthesis is developed without ground reaction in a first case and introduces ground effect in a second one to ameliorate prothesis performances. The obtained models are used to synthesize a control strategy based on TS fuzzy concepts and PID control to reproduce human lower limb behavior in a normal gait and climb and descent of stairs. The RSME errors are calculated to evaluate and compare the various results performances and eventually show the capacity of the proposed control with ground reaction impact on trajectory tracking. The RMSE values obtained for the four outputs of the fuzzy controller are very small for the different modes of locomotion. Moreover, they become weaker when the ground reaction forces are added to the model to show the role of these forces for the body equilibrium maintaining during the gait cycle. The developed approach ensured good trajectories tracking compared to a healthy leg even in presence of disturbances.

Manoeuvring speed estimation of a lane-change system using geometric hermite interpolation

Autonomous vehicles demand a continuous path to provide the passengers with physical ease and relaxation. Smooth experience in autonomous vehicles is very essential and can be accomplished by having good trajectories. The trajectory generated by the interpolation algorithm for Geometric Hermite is provided in this paper. The proposed algorithm calculates a path based on a set of geometric Hermite data that consists of predefined points including prescribed tangent angles and curvature values. The generated path has the potential to determine the maximum speed allowed for safe autonomous driving on roads by providing curvature information. It describes an alternative way to estimate design speed permitted on the road for autonomous lane change systems.

Asynchronous Episodic Deep Deterministic Policy Gradient: Toward Continuous Control in Computationally Complex Environments.

Deep deterministic policy gradient (DDPG) has been proved to be a successful reinforcement learning (RL) algorithm for continuous control tasks. However, DDPG still suffers from data insufficiency and training inefficiency, especially, in computationally complex environments. In this article, we propose asynchronous episodic DDPG (AE-DDPG), as an expansion of DDPG, which can achieve more effective learning with less training time required. First, we design a modified scheme for data collection in an asynchronous fashion. Generally, for asynchronous RL algorithms, sample efficiency or/and training stability diminish as the degree of parallelism increases. We consider this problem from the perspectives of both data generation and data utilization. In detail, we redesign experience replay by introducing the idea of episodic control so that the agent can latch on good trajectories rapidly. In addition, we also inject a new type of noise in action space to enrich the exploration behaviors. Experiments demonstrate that our AE-DDPG achieves higher rewards and requires less time consumption than most popular RL algorithms in learning to run task which has a computationally complex environment. Not limited to the control tasks in the computationally complex environments, AE-DDPG also achieves higher rewards and two-fold to four-fold improvement in sample efficiency on average compared with other variants of DDPG in MuJoCo environments. Furthermore, we verify the effectiveness of each proposed technique component through abundant ablation study.

Gestures In-The-Wild: Detecting Conversational Hand Gestures in Crowded Scenes Using a Multimodal Fusion of Bags of Video Trajectories and Body Worn Acceleration

This paper addresses the detection of hand gestures during free-standing conversations in crowded mingle scenarios. Unlike the scenarios of the previous works in gesture detection and recognition, crowded mingle scenes have additional challenges such as cross-contamination between subjects, strong occlusions, and nonstationary backgrounds. This makes them more complex to analyze using computer vision techniques alone. We propose a multimodal approach using video and wearable acceleration data recorded via smart badges hung around the neck. In the video modality, we propose to treat noisy dense trajectories as bags-of-trajectories. For a given bag, we can have good trajectories corresponding to the subject, and bad trajectories due for instance to cross-contamination. However, we hypothesize that for a given class, it should be possible to learn trajectories that are discriminative while ignoring noisy trajectories. We do this by exploiting multiple instance learning via embedded instance selection as our multiple instance learning approach. This technique also allows us to identify which instances contribute more to the classification. By fusing the decisions of the classifiers from the video and wearable acceleration modalities, we show improvements over the unimodal approaches with an AUC of 0.69. We also present a static analysis and a dynamic analysis to assess the impact of noisy data on the fused detection results, showing that the moments of high occlusion in the video are compensated by the information from the wearables. Finally, we applied our method to detect speaking status, leveraging the close relationship found in the literature between hand gestures and speech.

Evaluation of Changes in Functional Status in the Year After Aortic Valve Replacement

Functional status is a patient-centered outcome that is important for a meaningful gain in health-related quality of life after aortic valve replacement. To determine functional status trajectories in the year after transcatheter aortic valve replacement (TAVR) and surgical aortic valve replacement (SAVR). A prospective cohort study with a 12-month follow-up was conducted at a single academic center in 246 patients undergoing TAVR or SAVR for severe aortic stenosis. The study was conducted between February 1, 2014, and June 30, 2017; data analysis was performed from December 27, 2017, to May 7, 2018. Preoperative comprehensive geriatric assessment was performed and a deficit-accumulation frailty index (CGA-FI) (range, 0-1; higher values indicate greater frailty) was calculated. Telephone interviews were conducted to assess self-reported ability to perform 22 activities and physical tasks at 1, 3, 6, 9, and 12 months after the procedure. Of the 246 patients included in the study, 143 underwent TAVR (74 [51.7%] women; mean [SD] age, 84.2 [5.9] years), and 103 underwent SAVR (46 [44.7%] women; age, 78.1 [5.3] years). Five trajectories were identified based on functional status at baseline and during the follow-up: from excellent at baseline to improvement at follow-up (excellent baseline-improvement), good (high baseline-full recovery), fair (moderate baseline-minimal decline), poor (low baseline-moderate decline), and very poor (low baseline-large decline). After TAVR, the most common trajectory was fair (54 [37.8%]), followed by good (33 [23.1%]), poor (21 [14.7%]), excellent (20 [14.0%]), and very poor (12 [8.4%]) trajectories. After SAVR, the most common trajectory was good (39 [37.9%]), followed by excellent (38 [36.9%]), fair (20 [19.4%]), poor (3 [2.9%]), and very poor (1 [1.0%]) trajectories. Preoperative frailty level was associated with lower probability of functional improvement and greater probability of functional decline. After TAVR, patients with CGA-FI level of 0.20 or lower had excellent (3 [50.0%]) or good (3 [50.0%]) trajectories, whereas most patients with CGA-FI level of 0.51 or higher had poor (10 [45.5%]) or very poor (5 [22.7%]) trajectories. After SAVR, most patients with CGA-FI level of 0.20 or lower had excellent (24 [58.5%]) or good (15 [36.6%]) trajectories compared with a fair trajectory (5 [71.4%]) in those with CGA-FI levels of 0.41 to 0.50. Postoperative delirium and major complications were associated with functional decline after TAVR (delirium present vs absent: 14 [50.0%] vs 11 [13.4%]; complications present vs absent: 14 [51.9%] vs 19 [16.4%]) or lack of improvement after SAVR (delirium present vs absent: 27 [69.2%] vs 31 [81.6%]; complications present vs absent: 10 [62.5%] vs 69 [79.3%]). The findings suggest that functional decline or lack of improvement is common in older adults with severe frailty undergoing TAVR or SAVR. Although this nonrandomized study does not allow comparison of the effectiveness between TAVR and SAVR, anticipated functional trajectories may inform patient-centered decision making and perioperative care to optimize functional outcomes.

A Trajectory Planning Model for the Manipulation of Particles in Microfluidics

Many important microfluid applications require the control and transport of particles immersed in a fluid. We propose a model for automatically planning good trajectories from an arbitrary point to a target in the presence of obstacles. It can be used for the manipulation of particles using actuators of mechanical or electrical type. We present the mathematical formulation of the model and a numerical method based on the optimization of travel time through the Bellman's principle. The implementation is focused on square grids such as those built from pixelated images. Numerical simulations show that the trajectory tree produced by the algorithm successfully avoids obstacles and stagnant regions of the fluid domain.

Distributed wireless power transfer in sensor networks with multiple Mobile Chargers

We investigate the problem of efficient wireless power transfer in wireless sensor networks. In our approach, special mobile entities (called the Mobile Chargers) traverse the network and wirelessly replenish the energy of sensor nodes. In contrast to most current approaches, we envision methods that are distributed and use limited network information. We propose four new protocols for efficient charging, addressing key issues which we identify, most notably (i) what are good coordination procedures for the Mobile Chargers and (ii) what are good trajectories for the Mobile Chargers. Two of our protocols (DC, DCLK) perform distributed, limited network knowledge coordination and charging, while two others (CC, CCGK) perform centralized, global network knowledge coordination and charging. As detailed simulations demonstrate, one of our distributed protocols outperforms a known state of the art method, while its performance gets quite close to the performance of the powerful centralized global knowledge method.

Wireless energy transfer in sensor networks with adaptive, limited knowledge protocols

We investigate the problem of efficient wireless energy transfer in Wireless Rechargeable Sensor Networks (WRSNs). In such networks a special mobile entity (called the Mobile Charger) traverses the network and wirelessly replenishes the energy of sensor nodes. In contrast to most current approaches, we envision methods that are distributed, adaptive and use limited network information. We propose three new, alternative protocols for efficient charging, addressing key issues which we identify, most notably (i) to what extent each sensor should be charged, (ii) what is the best split of the total energy between the charger and the sensors and (iii) what are good trajectories the Mobile Charger should follow. One of our protocols (LRP) performs some distributed, limited sampling of the network status, while another one (RTP) reactively adapts to energy shortage alerts judiciously spread in the network. We conduct detailed simulations in uniform and non-uniform network deployments, using three different underlying routing protocol families. In most cases, both our charging protocols significantly outperform known state of the art methods, while their performance gets quite close to the performance of the global knowledge method (GKP) we also provide.

Modeling pubertal timing and tempo and examining links to behavior problems.

Research on the role of puberty in adolescent psychological development requires attention to the meaning and measurement of pubertal development. Particular questions concern the utility of self-report, the need for complex models to describe pubertal development, the psychological significance of pubertal timing vs. tempo, and sex differences in the nature and psychological significance of pubertal development. We used longitudinal self-report data to model linear and logistic trajectories of pubertal development, and used timing and tempo estimates from these models, and from traditional approaches (age at menarche and time from onset of breast development to menarche), to predict psychological outcomes of internalizing and externalizing behavior problems, and early sexual activity. Participants (738 girls, 781 boys) reported annually from ages 9 through 15 on their pubertal development, and they and their parents reported on their behavior in mid-to-late adolescence and early adulthood. Self-reports of pubertal development provided meaningful data for both boys and girls, producing good trajectories, and estimates of individuals' pubertal timing and tempo. A logistic model best fit the group data. Pubertal timing was estimated to be earlier in the logistic compared to linear model, but linear, logistic, and traditional estimates of pubertal timing correlated highly with each other and similarly with psychological outcomes. Pubertal tempo was not consistently estimated, and associations of tempo with timing and with behavior were model dependent. Advances in modeling facilitate the study of some questions about pubertal development, but assumptions of the models affect their utility in psychological studies.

Dynamic analogue initialization for ensemble forecasting

This paper introduces a new approach for the initialization of ensemble numerical forecasting: Dynamic Analogue Initialization (DAI). DAI assumes that the best model state trajectories for the past provide the initial conditions for the best forecasts in the future. As such, DAI performs the ensemble forecast using the best analogues from a full size ensemble. As a pilot study, the Lorenz63 and Lorenz96 models were used to test DAI’s effectiveness independently. Results showed that DAI can improve the forecast significantly. Especially in lower-dimensional systems, DAI can reduce the forecast RMSE by ∼50% compared to the Monte Carlo forecast (MC). This improvement is because DAI is able to recognize the direction of the analysis error through the embedding process and therefore selects those good trajectories with reduced initial error. Meanwhile, a potential improvement of DAI is also proposed, and that is to find the optimal range of embedding time based on the error’s growing speed.

General Path Planning Methodology for Leader-Follower Robot Formations

This paper describes a robust algorithm for mobile robot formations based on the Voronoi Fast Marching path planning method. This is based on the propagation of a wave throughout the model of the environment, the wave expanding faster as the wave's distance from obstacles increases. This method provides smooth and safe trajectories and its computational efficiency allows us to maintain a good response time. The proposed method is based on a local-minima-free planner; it is complete and has an O(n) complexity order where n is the number of cells of the map. Simulation results show that the proposed algorithm generates good trajectories.

Effects of different potential functions on modeling of RO membrane performance by use of an advanced model

Modeling of reverse osmosis membranes is essential for progress in membrane separation technology. Effects of a new potential function on the modified surface force pore flow (MD-SF-PF) model were investigated by Moradi et al. in 2012. The main purpose of this research was to study the effects of other potential functions on the advanced and extended form of the MD-SF-PF model (i.e. the EX-MD-SF-PF model). The results obtained were compared by use of an objective function and by using the root-mean-square deviation. The best potential function was finally selected and a new model was proposed. The complex and nonlinear equations of the EX-MD-SF-PF model were solved by use of advanced numerical methods, for example finite difference and the finite volume. The proposed model, besides enabling excellent prediction of the performance of membranes, for example separation factor, and solvent and pure fluxes, has much better root-mean-square deviation than the EX-MD-SF-PF model. Profiles of potential function against pore radius and pore length were plotted and good trajectories were obtained. Profiles of solute concentration at the beginning, middle, and end of a pore were calculated and compared. The velocity profile of the permeating solvent was calculated and plotted. Most of the experimental data were predicted by use of the model.

Symbolic approximate time-optimal control

There is an increasing demand for controller design techniques capable of addressing the complex requirements of today’s embedded applications. This demand has sparked the interest in symbolic control where lower complexity models of control systems are used to cater for complex specifications given by temporal logics, regular languages, or automata. These specification mechanisms can be regarded as qualitative since they divide the trajectories of the plant into bad trajectories (those that need to be avoided) and good trajectories. However, many applications require also the optimization of quantitative measures of the trajectories retained by the controller, as specified by a cost or utility function. As a first step towards the synthesis of controllers reconciling both qualitative and quantitative specifications, we investigate in this paper the use of symbolic models for time-optimal controller synthesis. We consider systems related by approximate (alternating) simulation relations and show how such relations enable the transfer of time-optimality information between the systems. We then use this insight to synthesize approximately time-optimal controllers for a control system by working with a lower complexity symbolic model. The resulting approximately time-optimal controllers are equipped with upper and lower bounds for the time to reach a target, describing the quality of the controller. The results described in this paper were implemented in the Matlab Toolbox Pessoa (Mazo et al., 2009 [12]) which we used to workout several illustrative examples reported in this paper.

Is there a caustic crossing in the lensed quasar Q2237+0305 observational data record?

We re-investigate the gravitationally lensed system Q2237+0305 data record to quantify the probability of having a caustic crossing in the A component. Several works assume that this is the case, but no quantitative analysis is available in the literature. We combine the data sets from the OGLE and GLITP collaborations to accurately trace the prominent event in the light curve for the A component of the system. Then the observed event is compared with synthetic light curves derived from trajectories in magnification maps. These maps are generated using a ray-tracing technique. We take more than 109 trajectories and test a wide range of different physical properties of the lensing galaxy and the source quasar (lens transverse velocity, microlens mass, source intensity profile and source size). We found that around 75 per cent of our good trajectories (i.e. that are consistent with the observations) are caustic crossings. In addition, a high transverse velocity exceeding 300 km s−1, a microlens mass of about 0.1 M⊙ and a small standard accretion disc is the best parameter combination. The results justify the interpretation of the OGLE–GLITP event in Q2237+0305A as a caustic crossing. Moreover, the physical properties of the lens and source are in very good agreement with previous works. We also remark that a standard accretion disc is preferred to those simpler approaches, and that the former should be used in subsequent simulations.