Explore the vast range of titles available.

Electric-vehicle technology is an emerging area offering several benefits such as economy due to low running costs. Electric vehicles can also help to significantly reduce CO2 emission, which is a vital factor for environmental pollution. Modern vehicles are equipped with driver-assistance systems that facilitate drivers by offloading some of the tasks a driver does while driving. Human beings are prone to errors. Therefore, accidents and fatalities can happen if the driver fails to perform a particular task within the deadline. In electric vehicles, the focus has always been to optimize the power and battery life, and thus, any additional hardware can affect their battery life significantly. In this paper, the design of driver-assistance systems has been introduced to automate and assist in some of the vital tasks, such as a braking system, in an optimized manner. We revamp the idea of the traditional driver-assistance system and propose a generic lightweight system based on the leading factors and their impact on accidents. We model tasks for these factors and simulate a low-cost driver-assistance system in a real-time context, where these scenarios are investigated and tasks schedulability is formally proved before deploying them in electric vehicles. The proposed driver-assistance system offers many advantages. It decreases the risk of accidents and monitors the safety of driving. If, at some point, the risk index is above a certain threshold, an automated control algorithm is triggered to reduce it by activating different actuators. At the same time, it is lightweight and does not require any dedicated hardware, which in turn has a significant advantage in terms of battery life. Results show that the proposed system not only is accurate but also has a very negligible effect on energy consumption and battery life.

This study investigates the nonlinear dynamics of Brushless DC (BLDC) motors using the MATLAB/Simulink platform, emphasizing system identification through the Least Squares (LS) method and Nonlinear Autoregressive Network With Exogenous Inputs (NARX) models with variable regressors. Accurate data-driven models derived from these techniques are essential for designing efficient feedback control systems, enabling precise motor dynamics representation and facilitating early fault detection by identifying deviations from normal operation. A detailed simulation of the BLDC motor under no-load conditions is performed to analyze the speed response and ripple effects in torque and speed, underscoring the need for effective modeling and control. Comprehensive datasets are generated to develop LS and NARX models across varying operational conditions. A variable step input voltage signal with both ascending and descending steps is employed for training, while a distinct validation signal of similar trends but different magnitudes is used for performance evaluation. All techniques are benchmarked using identical training and validation signals. Among the models, the NARX model with customized regressors demonstrated superior performance, achieving 99.1% training accuracy and 98.01% validation accuracy in predicting motor dynamics. All the models are further tested under real-time signal conditions like ramp-up acceleration, deceleration, & turning and noisy signal conditions to evaluate the robustness and accuracy of these models in real-world conditions. The findings highlight the NARX model’s potential to enhance control strategies and improve BLDC motor stability, with statistical analysis confirming the robustness and effectiveness of the proposed approach.

Advances in visual perceptual tasks have been mainly driven by the amount, and types, of annotations of large-scale datasets. Researchers have focused on fully-supervised settings to train models using offline epoch-based schemes. Despite the evident advancements, limitations and cost of manually annotated datasets have hindered further development for event perceptual tasks, such as detection and localization of objects and events in videos. The problem is more apparent in zoological applications due to the scarcity of annotations and length of videos-most videos are at most ten minutes long. Inspired by cognitive theories, we present a self-supervised perceptual prediction framework to tackle the problem of temporal event segmentation by building a stable representation of event-related objects. The approach is simple but effective. We rely on LSTM predictions of high-level features computed by a standard deep learning backbone. For spatial segmentation, the stable representation of the object is used by an attention mechanism to filter the input features before the prediction step. The self-learned attention maps effectively localize the object as a side effect of perceptual prediction. We demonstrate our approach on long videos from continuous wildlife video monitoring, spanning multiple days at 25 FPS. We aim to facilitate automated ethogramming by detecting and localizing events without the need for labels. Our approach is trained in an online manner on streaming input and requires only a single pass through the video, with no separate training set. Given the lack of long and realistic (includes real-world challenges) datasets, we introduce a new wildlife video dataset–nest monitoring of the Kagu (a flightless bird from New Caledonia)–to benchmark our approach. Our dataset features a video from 10 days (over 23 million frames) of continuous monitoring of the Kagu in its natural habitat. We annotate every frame with bounding boxes and event labels. Additionally, each frame is annotated with time-of-day and illumination conditions. We will make the dataset, which is the first of its kind, and the code available to the research community. We find that the approach significantly outperforms other self-supervised, traditional (e.g., Optical Flow, Background Subtraction) and NN-based (e.g., PA-DPC, DINO, iBOT), baselines and performs on par with supervised boundary detection approaches (i.e., PC). At a recall rate of 80%, our best performing model detects one false positive activity every 50 min of training. On average, we at least double the performance of self-supervised approaches for spatial segmentation. Additionally, we show that our approach is robust to various environmental conditions (e.g., moving shadows). We also benchmark the framework on other datasets (i.e., Kinetics-GEBD, TAPOS) from different domains to demonstrate its generalizability. The data and code are available on our project page: https://aix.eng.usf.edu/research_automated_ethogramming.html

Consider learning tasks where the precision requirement is very high, for example a 99 % precision requirement for a video classification application. We report that when very different sources of evidence such as text, audio, and video features are available, combining the outputs of base classifiers trained on each feature type separately, aka late fusion, can substantially increase the recall of the combination at high precisions, compared to the performance of a single classifier trained on all the feature types, i.e., early fusion, or compared to the individual base classifiers. We show how the probability of a joint false-positive mistake can be less—in some cases significantly less—than the product of individual probabilities of conditional false-positive mistakes (a NoisyOR combination). Our analysis highlights a simple key criterion for this boosted precision phenomenon and justifies referring to such feature families as (nearly) independent. We assess the relevant factors for achieving high precision empirically, and explore combination techniques informed by the analysis.

We propose a numerical-based approach extending the conditional MVAR Granger causality (MVGC) analysis for the construction of directed connectivity networks in the presence of both exogenous/stimuli and modulatory inputs. The performance of the proposed scheme is validated using both synthetic stochastic data considering also the influence of haemodynamics latencies and a benchmark fMRI dataset related to the role of attention in the perception of visual motion. The particular fMRI dataset has been used in many studies to evaluate alternative model hypotheses using the Dynamic Causal Modelling (DCM) approach. Based on the use of the Bayes factor, we show that the obtained GC connectivity network compares well to a reference model that has been selected through DCM analysis among other candidate models. Thus, our findings suggest that the proposed scheme can be successfully used as a stand-alone or complementary to DCM approach to find directed causal connectivity patterns in task-related fMRI studies.

Real-world autonomous systems operate under uncertainty about both their pose and dynamics. Autonomous control systems must simultaneously perform estimation and control tasks to maintain robustness to changing dynamics or modelling errors. However, information gathering actions often conflict with optimal actions for reaching control objectives, requiring a trade-off between exploration and exploitation. The specific problem setting considered here is for discrete-time non-linear systems, with process noise, input-constraints, and parameter uncertainty. This study frames this problem as a Bayes-adaptive Markov decision process and solves it online using Monte Carlo tree search with an unscented Kalman filter to account for process noise and parameter uncertainty. This method is compared with certainty equivalent model predictive control and a tree search method that approximates the QMDP solution, providing insight into when information gathering is useful. Discrete time simulations characterise performance over a range of process noise and bounds on unknown parameters. An offline optimisation method is used to select the Monte Carlo tree search parameters without hand-tuning. In lieu of recursive feasibility guarantees, a probabilistic bounding heuristic is offered that increases the probability of keeping the state within a desired region.

A single-degree-of-freedom nonlinear spring–mass system, subjected to a particular shaped input force whose magnitude varies with time in a piecewise-constant manner is considered. The goal is to bring the point mass in the model system from initial rest to a prescribed new equilibrium position without exciting any residual vibration. If, ideally, the potential energy associated with the elastic spring is known as a function of its elongation, the magnitude and execution time of each force step that serve the abovesaid purpose can be calculated by analyzing the mechanical energy flow. However, in practice the potential function almost inevitably contains a small estimation error, and residual vibration would be excited by the input force so calculated. By use of asymptotic techniques, the residual vibration excited by a two-step input force with slightly incorrect task time and force magnitudes is calculated. It is also demonstrated that, by comparing the closed-form results of the asymptotic analysis with online measurements of the excited residual vibration, the shape of the two-step input force (characterized by the task time and force magnitudes) can be corrected iteratively, thereby suppressing the residual vibration.

We propose and develop a discrete-time, continuous-flow model for studying the operation of a job shop that sees a stationary input mix of job types. We are not concerned with issues of detailed scheduling, but rather hope to develop a tactical planning tool for a job-shop operation. With the model, we are able to characterize the operational behavior of each work center in the job shop for a given control policy. The control rule sets the production rate at a work center as a fixed proportion of its queue level in each time period, and is consistent with the assignment of a planned lead time to each work center. For these control rules, the model gives the steady-state distribution of the production levels at each work center, as well as the distribution of queue lengths. We show how to use the model not only to evaluate a choice of the controls but also to find a good specification that results in acceptable shop behavior. An example for a factory that produces grinding machines illustrates the use of the model.