Task Cost Research Articles

Boosting fault localization of statements by combining topic modeling and Ochiai

Context:Reducing the cost of maintenance tasks by fixing bugs automatically is the cornerstone of Automated Program Repair (APR). To do this, automated Fault Localization (FL) is essential. Two families of FL techniques are Spectrum-based Fault Localization (SBFL) and Information Retrieval Fault Localization (IRFL). In SBFL, the coverage information and execution results of test cases are utilized. Ochiai is one of the most effective and used SBFL strategies. In IRFL, the bug report information is utilized as well as the identifier names and comments in source code files. Latent Dirichlet Allocation (LDA) is a generative statistical model and one of the most popular topic modeling methods. However, LDA has been used at the method level of granularity as IRFL technique, whereas most existing APR tools are focused on the statement level. Objective:This paper presents our approach that combines topic modeling and Ochiai to boost FL at the statement level. Method:We evaluate our approach considering five different projects in Defects4J benchmark. We report the performance of our approach in terms of hit@k and MRR. To study the impact on the results, we compare our approach against five baselines: two SBFL approaches (Ochiai and Dstar), two IRFL approaches (LDA and Blues), and one hybrid approach (SBIR). In addition, we compare the number of bugs that are found by our approach with the baselines. Results:Our approach significantly outperforms the baselines in all metrics. Especially, when hit@1, hit@3 and hit@5 are compared. Also, our approach locates more bugs than Ochiai and Blues. Conclusion:The results of our approach indicate that the integration of topic modeling with Ochiai boosts FL. This uncovers the potential of topic modeling for FL at statement level, which is valuable for the APR community.

Controlling response order without relying on stimulus order – evidence for flexible representations of task order

In dual-task situations, both component tasks are typically not executed simultaneously but rather one after another. Task order is usually determined based on bottom-up information provided by stimulus presentation order, but also affected by top-down factors such as instructions and/or differentially dominant component tasks (e.g., oculomotor task prioritization). Recent research demonstrated that in the context of a randomly switching stimulus order, task order representations can be integrated with specific component task information rather than being coded in a purely abstract fashion (i.e., by containing only generic order information). This conclusion was derived from observing consistently smaller task-order switch costs for a preferred (e.g., oculomotor-manual) versus a non-preferred (e.g., manual-oculomotor) task order (i.e., order-switch cost asymmetries). Since such a representational format might have been especially promoted by the sequential stimulus presentation employed, we investigated task-order representations in situations without any bottom-up influence of stimulus order. To this end, we presented task stimuli simultaneously and cued the required task-order in advance. Experiment 1 employed abstract order transition cues that only indicated a task-order repetition (vs. switch) relative to the previous trial, while Experiment 2 used explicit cues that unambiguously indicated the task-order. Experiment 1 revealed significant task-order switch costs only for the second task (of either task order) and no order-switch cost asymmetries, indicating a rather generic representation of task order. Experiment 2 revealed task-order switch costs in both component tasks with a trend toward order-switch cost asymmetries, indicating an integration of task order representations with component task information. These findings highlight an astonishing flexibility of mental task-order representations during task-order control.

The Association Between Postural Sway and Preclinical Alzheimer Disease Among Community-Dwelling Older Adults.

It is unknown whether older adults with preclinical Alzheimer disease (AD) experience changes in postural sway compared with those without preclinical AD. The purpose of this study was to understand the effect of dual tasking on standing balance, or postural sway, for people with and without preclinical AD. A cross-sectional analysis of baseline data from a longitudinal cohort study. Participants were cognitively normal older adults with and without preclinical AD. Postural sway (path length) was tested using a force plate under standard and dual task balance conditions. Dual task cost (DTC) was calculated to examine performance change in balance conditions. Logistic regression models were used to predict preclinical AD status as a function of DTC. 203 participants (65 preclinical AD+) were included. DTC for path length was significantly greater for participants with preclinical AD (DTC path length mean difference 19.8, 95% CI 2.6-37.0, t(201) = 2.29, p = .024). Greater DTC was significantly associated with increased odds of having preclinical AD (adjusted odds ratio for a 20-unit increase in DTC 1.16, 95% CI 1.02-1.32). Older adults with preclinical AD are more likely to demonstrate significantly greater DTC in postural sway than those without preclinical AD. Dual tasking should be integrated into balance and fall risk assessments and may inform early detection of preclinical AD.

Cognitive function influences cognitive-motor interference during dual task walking in multiple sclerosis

BackgroundBoth physical and cognitive impairments are common in people with multiple sclerosis (PwMS). Performing a cognitive task while walking (i.e., dual-task walking) can introduce cognitive-motor interference (CMI), resulting in changes in walking performance. The association between the levels of cognitive impairment and of CMI in MS remains unclear. ObjectivesTo examine the association between cognitive functioning and differences in walking performance arise between single- and dual-task walking. MethodsNinety-five PwMS performed self-preferred pace walking and dual-task walking. The gait parameters recorded were used to compute dual task costs (DTC) as a metric of CMI. Cognitive functioning was assessed using Match, an unsupervised test developed based on the Symbol Digit Modalities Test. Participants were categorized as higher (HCF) and lower cognitive functioning (LCF) based on a Match z-score < −1.5. ResultsLCF group had elevated DTC for stride velocity, relative to the HCF group. Higher DTC for stride velocity was associated with lower cognition, as assessed by Match test. ConclusionThe findings support the hypothesis that CMI is associated with cognitive functioning in PwMS.

Effects of Task Interference on Kinematics and Dual-Task Cost of Running in Early Childhood.

Children aged 3-8 are in a critical period for motor development and postural control. Running is a basic motor skill that children need to master in early childhood. While running, children are prone to dangerous events such as falls. This study investigates the kinematic characteristics of running by children associated with different interference tasks, i.e., normalized running, cognitive dual-tasks, and obstacle crossing tasks, and provides a theoretical foundation for the interference mechanism of children's dynamic postural control and for screening of motor disorders. Two hundred children aged 3-8 were recruited. The BTS Bioengineering infrared motion capture system was used to collect spatiotemporal and kinematic running data under three tasks. Repeated measures of variance analysis were used to compare the effects of different interference tasks and ages on children's running signs. The main and interaction effect tests were compared by the Bonferroni method. The results and conclusions are as follows: (1) Running characteristics of early childhood are influenced by interference tasks and age. With interference tasks, the overall characteristics of running by children aged 3-8 showed an increasing trend in running cycle time and a decreasing trend in stride length, step length, cadence, and speed. (2) Both cognitive and obstacle crossing tasks had costs, and cognitive task costs were greater than obstacle crossing costs. Children adopted a "task first" running strategy with different interference tasks. When facing cognitive tasks, their overall joint motion decreased, and they reduced joint motions to promote task completion. When facing obstacle crossing tasks, because of the characteristics of the task itself, children increased joint motions to cope with interference. (3) In terms of age, the running characteristics showed a nonlinear development trend in various indicators, with a degree of recurrence and high variability in adjacent age groups. (4) The dual-task interference paradigm of "postural-cognition" can be used as a motor intervention tool to promote the development of basic motor skills in early childhood.

The effect of single and dual task training on the intracortical inhibition in healthy young adults

Introduction&#x0D; The evidence for changes in intracortical inhibition when executing two tasks simultaneously (i.e., dual tasking) is ambiguous as decreased (Corp et al., 2014) and increased (Corp et al., 2016) inhibition were reported. One way to bring more light into this question is to tests the effect of a single task training (STT) and a dual task training (DTT) on the short interval intracortical inhibition (SICI) during a single balancing task and two different dual tasks in healthy young adults.&#x0D; Methods&#x0D; Twenty-nine healthy young adults were randomly separated into two groups participating in STT (n = 15) or DTT (n = 14) consisting of 6 training sessions within 3 weeks. Before and after the training, a single task (balancing on a rocker board) was performed at two resistance levels (easy and hard). Additionally to the single task, either a cognitive (2-back number recall) or a motor (balancing a ball on a hand-held tray) dual task was executed simultaneously. During execution of these three tasks, SICI was measured with transcranial magnetic stimulation over the motor cortical area representing the right tibialis anterior.&#x0D; Results&#x0D; Training improvements in balance performance were group and task-specific over time (p = .018). While the STT group improved more in the single balance task (12.3% vs. 6.6% DTT), the DTT group had more sway reductions in the motor dual task condition (13.7% vs. 4.5% STT). Similar statistical outcome (p = .034) was observed for the dual task costs (DTC). There was a tendence for SICI (p = .075), mainly indicating higher increase in SICI for the DTT group in the motor dual task (16.0% vs. 5.8% STT). During the execution of the single balance task, the group-specific adaptations in SICI were less pronounced (13.7% DTT vs. 16.2% STT). When analyzing the SICI dual task difference (Δ) from single to dual task, SICI is altered group and task specific (p = .011). The DTT group could increase the dual task difference in SICI in the dual motor condition (Δ 3.2%), whereas the STT group had a decrease (Δ -9.6%).&#x0D; Discussion/Conclusion&#x0D; The results of this study show that DTT causes gains in balance performance and increases in SICI when the secondary task is also a motor task, but not when the second task is a cognitive one. STT is particularly beneficial in the single task. It is therefore assumed that intracortical inhibition is important during the simultaneous performance of two motor tasks, while intracortical inhibition was not modulated in a group-specific manner by the additional cognitive task.&#x0D; References&#x0D; Corp, D. T., Lum, J. A. G., Tooley, G. A., &amp; Pearce, A. J. (2014). Corticospinal activity during dual tasking: A systematic review and meta-analysis of TMS literature from 1995 to 2013. Neuroscience &amp; Biobehavioral Reviews, 43, 74-87. https://doi.org/10.1016/j.neubiorev.2014.03.017&#x0D; Corp, D. T., Rogers, M. A., Youssef, G. J., &amp; Pearce, A. J. (2016). The effect of dual-task difficulty on the inhibition of the motor cortex. Experimental Brain Research, 234, 443-452. https://doi.org/10.1007/s00221-015-4479-2

Hybrid representation-enhanced sampling for Bayesian active learning in musculoskeletal segmentation of lower extremities.

Manual annotations for training deep learning models in auto-segmentation are time-intensive. This study introduces a hybrid representation-enhanced sampling strategy that integrates both density and diversity criteria within an uncertainty-based Bayesian active learning (BAL) framework to reduce annotation efforts by selecting the most informative training samples. The experiments are performed on two lower extremity datasets of MRI and CT images, focusing on the segmentation of the femur, pelvis, sacrum, quadriceps femoris, hamstrings, adductors, sartorius, and iliopsoas, utilizing a U-net-based BAL framework. Our method selects uncertain samples with high density and diversity for manual revision, optimizing for maximal similarity to unlabeled instances and minimal similarity to existing training data. We assess the accuracy and efficiency using dice and a proposed metric called reduced annotation cost (RAC), respectively. We further evaluate the impact of various acquisition rules on BAL performance and design an ablation study for effectiveness estimation. In MRI and CT datasets, our method was superior or comparable to existing ones, achieving a 0.8% dice and 1.0% RAC increase in CT (statistically significant), and a 0.8% dice and 1.1% RAC increase in MRI (not statistically significant) in volume-wise acquisition. Our ablation study indicates that combining density and diversity criteria enhances the efficiency of BAL in musculoskeletal segmentation compared to using either criterion alone. Our sampling method is proven efficient in reducing annotation costs in image segmentation tasks. The combination of the proposed method and our BAL framework provides a semi-automatic way for efficient annotation of medical image datasets.

Uninterruptible Energy Production in Standalone Power Systems for Telecommunications

A continuously energy production is the most difficult and important feature to be guaranteed when designing a standalone power system for telecommunications. This paper focuses on a standalone power system for telecommunications already proposed in the literature, combining the most mature generators and technologies related to renewable energies and environment friendliness. It is important to know the operational problems, the best way to solve them and how to apply efficient maintenance. Reliability studies based on existent applications of renewable generators (some of them used in standalone power systems) help to know what to expect from their operational behaviour and to reduce time and costs in maintenance tasks. Notwithstanding their undoubted importance, they are not much recognised in literature, yet. The paper presents a brief review based on the few reliability reports about wind and photovoltaic systems available in the literature. Contributions and important aspects are discussed having in mind all the obstacles to achieve an uninterruptible power supply.

Effectiveness and cost of integrated cognitive and balance training for balance and falls in cerebellar ataxia: a blinded two-arm parallel group RCT.

In patients with cerebellar ataxia (CA), dual-tasking deteriorates the performance of one or both tasks. Evaluate the effects of 4 weeks of cognitive-coupled intensive balance training (CIBT) on dual-task cost, dynamic balance, disease severity, number of falls, quality of life, cognition and cost among patients with CA. This RCT compared CIBT (Group 1) to single-task training (Group 2) among 32 patients with CA. The intervention included either dual-task (CIBT) or single-task training for 4 weeks followed by 6 months of unsupervised home exercises. Dual-task timed up-and-go test (D-TUG) assessed dual-task cost of the physical and cognitive tasks. Assessment time points included baseline 1 (Week 0:T1), baseline 2 (Week 6:T2), post-intervention (Week 10:T3), and follow-up (Week 34:T4). Compared to single-task training CIBT improved the dual-task cost of physical task [MD -8.36 95% CI (-14.47 to -2.36, p < 0.01), dual-tasking ability [-6.93 (-13.16 to -0.70); p = 0.03] assessed using D-TUG, balance assessed using the scale for the assessment and rating of ataxia (SARAbal) [-2.03 (-4.04 to -0.19); p = 0.04], visual scores of the SOT (SOT-VIS) [-18.53 (-25.81 to -11.24, p ≤ 0.01] and maximal excursion [13.84 (4.65 to 23.03; p ≤ 0.01] of the Limits of Stability (LOS) in the forward direction and reaction time in both forward [-1.11 (-1.42 to -0.78); p < 0.01] and right [-0.18 (0.05 to 0.31); p < 0.01] directions following 4 weeks of training. CIBT did not have any additional benefits in reducing the number of falls, or improving disease severity, quality of life and cognition. The mean cost of intervention and healthcare costs for 7 months was HKD 33,380 for CIBT group and HKD 38,571 for single-task training group. We found some evidence to support the use of CIBT for improving the dual-tasking ability, dual-task cost of physical task and dynamic balance in CA. Future large fully-powered studies are needed to confirm this claim. https://clinicaltrials.gov/study/NCT04648501, identifier [Ref: NCT04648501].

Dual-Task Cost Effects on Static Posture Control Parameters and Choice Reaction Time in Individuals With and Without Intellectual Disabilities.

This study aimed to evaluate the effect of age and intellectual disability (ID) on postural balance parameters, dual-task cost (DTC), and choice reaction time (CRT). Fifty-eight individuals with ID and 55 peers without ID performed a postural stance balance task under two conditions: a single task with eyes open and dual task involving an additional cognitive task (light sequence). Four postural balance parameters (total displacement, total sway area, mediolateral, and anteroposterior dispersion), cost of the dual task ([DTC%] = [(single-task performance - dual-task performance)/single-task performance] × 100), and CRT were recorded, calculated, and analyzed. All postural control parameters reflected poorerperformance during the dual-task condition, nevertheless, DTC was significantly higher only in individuals with ID and only for the total sway area, F(1, 111) = 5.039, p = .027, and mediolateral dispersion, F(1, 111) = 6.576, p = .012. CRT was longer in individuals with ID compared with the individuals without ID, F(1, 111) = 94.979, p ≤ .001, while age did not have a significant effect on the DTC nor on the CRT, F(1, 111) = 0.074, p = .786. In conclusion, an additional cognitive task during the postural balance task had a detrimental effect on various postural balance parameters, leading to increased DTC in terms of total sway area, mediolateral dispersion, and prolonged CRT in individuals with ID.

Transferability-Guided Cross-Domain Cross-Task Transfer Learning.

We propose two novel transferability metrics fast optimal transport-based conditional entropy (F-OTCE) and joint correspondence OTCE (JC-OTCE) to evaluate how much the source model (task) can benefit the learning of the target task and to learn more generalizable representations for cross-domain cross-task transfer learning. Unlike the original OTCE metric that requires evaluating the empirical transferability on auxiliary tasks, our metrics are auxiliary-free such that they can be computed much more efficiently. Specifically, F-OTCE estimates transferability by first solving an optimal transport (OT) problem between source and target distributions and then uses the optimal coupling to compute the negative conditional entropy (NCE) between the source and target labels. It can also serve as an objective function to enhance downstream transfer learning tasks including model finetuning and domain generalization (DG). Meanwhile, JC-OTCE improves the transferability accuracy of F-OTCE by including label distances in the OT problem, though it incurs additional computation costs. Extensive experiments demonstrate that F-OTCE and JC-OTCE outperform state-of-the-art auxiliary-free metrics by 21.1% and 25.8% , respectively, in correlation coefficient with the ground-truth transfer accuracy. By eliminating the training cost of auxiliary tasks, the two metrics reduce the total computation time of the previous method from 43 min to 9.32 and 10.78 s, respectively, for a pair of tasks. When applied in the model finetuning and DG tasks, F-OTCE shows significant improvements in the transfer accuracy in few-shot classification experiments, with up to 4.41% and 2.34% accuracy gains, respectively.

Investigation of the relationship between upper extremity lateralization and dual task cost in ındividuals with hemiplegic cerebral palsy

Investigation of the relationship between upper extremity lateralization and dual task cost in ındividuals with hemiplegic cerebral palsy

Deep reinforcement learning‐based joint task offloading and resource allocation in multipath transmission vehicular networks

AbstractMobile edge computing (MEC) is increasingly being applied to task computation with the continuing increase of computing‐intensive applications in vehicular networks. The MEC can offload the tasks generated by the vehicle to the MEC server to mitigate the computational constraints on the mobile vehicle. However, most task offloading algorithms can not cope with the dynamic distribution of vehicles in vehicular networks. Moreover, most task offloading algorithms do not consider vehicular networks using multipath transmission schemes. Therefore, it remains a challenge to implement efficient joint task offloading and resource allocation (JTORA) algorithms in multipath transmission vehicular networks. In this paper, we introduce a novel cooperative transmission network architecture in multipath transmission vehicular networks. The computational tasks generated by the vehicles can be offloaded to MEC servers deployed near the base station or to cloud servers via multipath transmission. We formulate the task offloading and resource allocation problem as a multi‐objective optimization problem in a multipath transmission vehicular network. To achieve efficient task offloading and resource allocation under cooperative transmission network architecture, we propose a JTORA algorithm for multipath transmission vehicular networks based on deep reinforcement learning (DRL). JTORA converts the optimization problem into a DRL model and obtains a specific optimization policy through model training. Simulation results show that JTORA achieves lower average task delay and lower average task cost than other traditional heuristic algorithms.

Pick-up point recommendation strategy based on user incentive mechanism

In recent years, with the development of spatial crowdsourcing technology, online car-hailing, as a typical spatiotemporal crowdsourcing task application scenario, has attracted widespread attention. Existing researches on spatial crowdsourcing are mainly based on the coordinate positions of user and worker roles to achieve task allocation with the goal of maximum matching number or lowest cost. However, they ignores the problem of the selection of the pick-up point which needs to be solved in the actual scene of online car booking. This problem needs to take into account the four-dimensional coordinate positions of users, workers, pick-up point and destination. Based on this, this study designs a pick-up point recommendation strategy based on user incentive mechanism. Firstly, a new four-dimensional crowdsourcing model is established, which is closer to the practical application of crowdsourcing problem. Secondly, taking cost optimization as the index, a user incentive mechanism is designed to encourage users to walk to the appropriate pick-up point within a certain distance. Thirdly, a concept of forward rate is proposed to reduce the computation time. Some key factors, such as the maximum walking distance limit of users and task cost, are considered as the recommendation index for measuring the pick-up point. Then, an effective pick-up point recommendation strategy is designed based on this index. Experiments show that the strategy proposed in this article can achieve reasonable recommendation for pick-up points and improve the efficiency of drivers and reduce the total trip cost of orders to the greatest extent.

Accurate energy barriers for catalytic reaction pathways: an automatic training protocol for machine learning force fields

We introduce a training protocol for developing machine learning force fields (MLFFs), capable of accurately determining energy barriers in catalytic reaction pathways. The protocol is validated on the extensively explored hydrogenation of carbon dioxide to methanol over indium oxide. With the help of active learning, the final force field obtains energy barriers within 0.05 eV of Density Functional Theory. Thanks to the computational speedup, not only do we reduce the cost of routine in-silico catalytic tasks, but also find an alternative path for the previously established rate-limiting step, with a 40% reduction in activation energy. Furthermore, we illustrate the importance of finite temperature effects and compute free energy barriers. The transferability of the protocol is demonstrated on the experimentally relevant, yet unexplored, top-layer reduced indium oxide surface. The ability of MLFFs to enhance our understanding of extensively studied catalysts underscores the need for fast and accurate alternatives to direct ab-initio simulations.

Spatiotemporal gait parameter fluctuations in older adults affected by mild cognitive impairment: comparisons among three cognitive dual-task tests

BackgroundsGait disorder is associated with cognitive functional impairment, and this disturbance is more pronouncedly when performing additional cognitive tasks. Our study aimed to characterize gait disorders in mild cognitive impairment (MCI) under three dual tasks and determine the association between gait performance and cognitive function.MethodsA total of 260 participants were enrolled in this cross-sectional study and divided into MCI and cognitively normal control. Spatiotemporal and kinematic gait parameters (31 items) in single task and three dual tasks (serial 100-7, naming animals and words recall) were measured using a wearable sensor. Baseline characteristics of the two groups were balanced using propensity score matching. Important gait features were filtered using random forest method and LASSO regression and further described using logistic analysis.ResultsAfter matching, 106 participants with MCI and 106 normal controls were recruited. Top 5 gait features in random forest and 4 ~ 6 important features in LASSO regression were selected. Robust variables associating with cognitive function were temporal gait parameters. Participants with MCI exhibited decreased swing time and terminal swing, increased mid stance and variability of stride length compared with normal control. Subjects walked slower when performing an extra dual cognitive task. In the three dual tasks, words recall test exhibited more pronounced impact on gait regularity, velocity, and dual task cost than the other two cognitive tests.ConclusionGait assessment under dual task conditions, particularly in words recall test, using portable sensors could be useful as a complementary strategy for early detection of MCI.

Unraveling the unparalleled benefits of orienteering versus hiking on gait performance and cognition: A randomized clinical trial

MethodsPhysical activities and sports in nature offer opportunities for leisure, health, and well-being. A randomized clinical trial (RCT) compared the effects of training, dualtask orienteering and single-task hiking in independent older adults (n = 90) allocated into three groups: orienteering group (OG), hiking group (HG) and control group (CG). Timed Up and Go (TUG) tests, including a dual-task (TUG-DT), and cognitive tests were administered. Differences were analyzed by Generalized Estimating Equations (GEE) and the effect size (ES) was calculated. Unpaired and paired tests were used for the evaluation of basic differences and after 24 weeks of intervention. ResultsThis RCT highlights the interaction effects between groups and time. Only the OG in TUGDT reduced by 6 s, with change of 42% (ES=0.77). Likewise, Dual TUG + cognitive stops (TUG-DT+ CS) and Dual TUG + cognitive stops + cognitive errors (TUG-DT + CS + CE), reduced 7 (ES=0.85) and 8 (ES=0.90) seconds (<0.001). The task cost improved by 75% for OG and 24% for HG (p = 0.001), with an ES of 0.56 for OG. For cognition, there was a 49.2% increase (ES=2.00) for OG and 16.5% for HG in executive functions and visuospatial skills, a 31.4% increase (ES=1.64) for OG and 14.7% for HG in processing speed, and a 39.8% increase (ES=1.62) for OG and 6.5% for HG in memory. All interactions were <0.001. ConclusionThe orienteering demonstrated that has potential benefits in effectively improving gait performance and preventing cognitive decline in older adults. Brazilian Registry of Clinical Trials (ReBEC): RBR-99jmmpf.

A novel coverage-aware task allocation scheme in Cooperative Mobile Crowd Sensing

Task assignment is a key issue in Mobile Crowd Sensing, which affects the quality and cost of tasks. Most of the existing works mainly consider the scenario where workers can complete tasks independently, ignoring the requirement of cooperation with multiple workers for complex sensing tasks. The quality of the cooperation of workers’ teams needs to be evaluated, and the task coverage should be improved in Cooperative Mobile Crowd Sensing. Therefore, in this paper, we first introduce the team cohesion index by analyzing the impact of workers’ social networks on tasks and then propose a new measurement method to better evaluate the quality of workers’ teamwork. Then, to improve the task coverage, the task assignment is modeled as a multi-constraint optimization problem in view of several factors, including location, worker team cohesion, and skill coverage. The greedy approach and the heuristic genetic algorithm are combined to create the two-stage algorithm known as GGA. Finally, experiments are conducted based on real datasets, which verify the effectiveness of the team cohesion index measurement method and task allocation algorithm GGA.

A Novel Multi-Robot Task Assignment Scheme Based on a Multi-Angle K-Means Clustering Algorithm and a Two-Stage Load-Balancing Strategy

A practical task assignment is one of the core issues of a multi-robot system. In this paper, a multi-robot task assignment strategy based on load balancing is proposed to effectively balance and plan out the execution cost of each robot when it has a large number of working task points. Considering the variability of the execution task cost in practical situations with different task point categories, the multi-robot task assignment (MRTA) problem is transformed into a multiple traveling salesman problem (MTSP) using a multi-angle K-means clustering algorithm. To solve the problem of unbalanced cost consumed by each robot after clustering assignment, which leads to low efficiency of system completion, a two-stage load-balancing strategy is presented. The first stage of this strategy makes a large adjustment to the unbalanced task set, and the second stage achieves a small fine-tuning to the unbalanced task set. The experimental results show that the standard deviation of the cost ratio of each set decreases when four robots perform the task between 100 and 550 work points using the load-balancing strategy. The reduction in standard deviation is between 3.53% and 83.44%. The maximum cost of individual robots decreases between 0.18% and 14.27%. The proposed method can effectively solve the uneven execution cost of each robot in the task assignment process and improve the efficiency of the system in completing tasks.

Unsupervised Retrieval Based Multivariate Time Series Anomaly Detection and Diagnosis with Deep Binary Coding Models

Retrieval based multivariate time series anomaly detection and diagnosis refer to identifying abnormal status in certain time steps and pinpointing the root cause input variables, i.e., sensors, by comparing a current time series segment and its relevant ones that are retrieved from huge amount of historical data. Binary coding with a deep neural network can be applied to reduce the computational cost of the retrieval tasks. However, it is hard to pinpoint the root cause sensors that are responsible for the anomaly, once multivariate time series segments are transformed into binary codes. In this paper, we present an unsupervised retrieval based multivariate time series anomaly detection and diagnosis method with deep binary coding model, to secure both efficiency and explainability. Specifically, we first transform input multivariate time series segments into low dimensional features with a temporal encoder. Subsequently, two hash functions predict two binary codes with different lengths from each feature. The binary codes with two different lengths can contribute to accelerate both anomaly detection and anomaly diagnosis. Experiments performed on datasets from various domains including real optical network, demonstrate the effectiveness and efficiency of the proposed method.