Adding Robustness in Dynamic Preemptive Scheduling

In this paper we present algorithms for jointly scheduling periodic tasks and hard aperiodic tasks with precedence relations in a distributed systems under the earliest deadline first algorithm. The periodic distributed processes are assumed to be guaranteed off-line and decomposed into a set of independent tasks by means of release time and deadline modifications. This paper proposes a similar on-line decomposition of aperiodic tasks using a guarantee algorithm based on dynamically-calculated response times. An extension of a dynamic slack stealing algorithm is proposed that allows to obtain minimal response times for aperiodic task with release time constraints. Based on such extension, an on-line algorithm splits every aperiodic distributed process into a set of independent tasks, determining the end-to-end response time and, therefore, allowing the use of hard aperiodic distributed processes.

A very important challenge in many virtual applications is to plan feasible, smooth and congestion-free paths for virtual agents in dynamic and complex environments. The agents should move towards their destinations successfully while avoiding the collisions with other agents and static and dynamic obstacles. In this paper, we propose a novel approach for realistic path planning. We first create a navigation mesh for the walkable regions in a two-dimensional environment. Then an A* search on this graph determines a series of connected meshes for agents to go through from the start position to the goal position and furthermore, the walkable corridor whose radii equal to maximum clearance to the obstacles is built based on backbone path derived from the inflection point method and Catmull-Rom spline. Finally, a local collision avoidance algorithm is integrated to guarantee that agents navigating in the corridor do not collide with other agents and dynamic obstacles. Our experiments show that we can compute feasible, smooth and realistic paths for agents situated in dynamic environments in real time.

The method of the current acceleration range navigation satellite signal simulator detection theoretically analyzed, but in the actual test work, how to set the satellite navigation signal simulator motion scenes and satellite receivers, a smaller level of the RF test signal navigation As well as implementing automated testing, there is a lack of practical and feasible testing methods. This paper simulates the RF signals received by receivers in different dynamic environments in real time, builds a dynamic range test staging, forms a specific test plan, and carries out the validity, practicability and effectiveness of the test verification test method. This test method is simple and reliable. The dynamic range of the satellite navigation signal simulator in the simulation environment can be accurately detected by this method.

Collective adaptive systems (CAS) have been attracting increasing attention in the field of artificial intelligence (AI), in which collaboration of agents plays a key role. These systems aim to accomplish a certain goal though collaborating between a variety of agents with different tasks, which adapt to changes of environment to be of adaptability. To solve the issue of collaboration of agents in an uncertain and highly dynamic environment, our research team had proposed a Goal-Capability-Commitment (GCC) based mediation for multi-agent collaboration, which generates the collaboration planning driven by capability based on global context states in real-time dynamic environment. As a case study for the application of GCC model, this paper adopts GCC to model the RoboCup Rescue Simulation System (RCRSS). As the result of modelling, the GCC domain model is applied to RCRSS where an efficiently quantitative evaluation is provided.

Robots are frequently presented with vast arrays of diverse data. Unfortunately, perfect memory and recall provides a mixed blessing. While flawless recollection of episodic data allows increased reasoning, photographic memory can hinder a robot's ability to operate in real-time dynamic environments. Human-inspired forgetting methods may enable robotic systems to rid themselves of out-dated, irrelevant, and erroneous data. This paper presents the use of human-inspired forgetting to act as a filter, removing unnecessary, erroneous, and out-of-date information. The novel ActSimple forgetting algorithm has been developed specifically to provide effective forgetting capabilities to robotic systems. This paper presents the ActSimple algorithm and how it was optimized and tested in a WiFi signal strength estimation task. The results generated by real-world testing suggest that human-inspired forgetting is an effective means of improving the ability of mobile robots to move and operate within complex and dynamic environments.

This paper presents a hard real-time kernel, called HARTIK, specifically designed to handle robotics applications with predictable response time. The main relevant features of this kernel include: direct specification of time constraints, such as periods and deadlines; preemptive scheduling; coexistence of hard soft, and non real-time tasks, separation between time constraints and importance; deadline tolerance; dynamic guarantee of critical tasks; and graceful degradation in overload conditions. The functionality of the kernel is then shown by presenting a concrete example of a robot system that has to explore unknown objects by visual and force feedback. >

One key prerequisite for a machine to interact intelligently with people is its ability to recognize humans as interaction partners, and to understand their behaviors and the intentions and plans behind them. In this paper we are particularly interested in behaviors which are related to motion, e.g. intentionally moving towards a goal or obstructing somebody's path. More specifically, we address the problem of detecting and tracking people in natural dynamic environments in realtime, and extracting and classifying typical motion patterns. Robust detection and tracking of nonrigid objects in natural changing environments with a moving observer is a challenging problem. Although various sophisticated approaches have been presented, robustness with respect to changing light conditions, non-rigidness of the tracked objects, occlusion and motion of the observer is still a problem which needs for an effective solution. Recognizing and understanding the intention of motion would leverage applications in areas such as smart environments, human-machine interaction and video surveillance. We present an approach for tracking people by integrating range information and visual information, and using their complimentary strengths to overcome typical tracking problems. Individual motion trajectories are extracted from these fused sensor data and are condensed in compact statistical spatiotemporal models. These models serve as a basis for the extraction of typical motion patterns by unsupervised learning.

Tracking of moving objects in video streams with considering different dynamic backgrounds is a challenging problem in a real time dynamic environment. The proposed approach uses an observation model based on optical flow information used to know the displacement of the objects present in the scene. After finding the moving regions in the initial frame, we are applying active contour model (ACM) to track the moving objects in the further frames dynamically. These models have been used as a natural means of incorporating flow information into the tracking. The formulation of the active contour model by incorporating an additional force driven optical flow field improves the tracking speed. This algorithm efficiently works to track for low contrast videos like aerial videos.

Collision Avoidance Systems (CASs) are attracting a lot of attention as one of the most preferred solutions for advanced driver assistance and autonomous driving. However, scene understanding, which is an essential functionality in CASs, remains a major challenge mainly due to the need for real-time understanding of highly dynamic and complex environment. In this research, a number of robust and low complexity vision based scene understanding techniques for collision avoidance on roadway have been proposed. It has been well recognized in the literature that road surface detection in a dynamic environment is both challenging and computationally intensive. An efficient non-parametric road surface detection algorithm that exploits the depth cue is proposed to overcome the limitations of existing road surface detection methods. Unlike existing methods that attempt to fit the road surface into rigid models, the proposed method results in low computational complexity, mainly due to the reliance on four intrinsic road scene attributes observed under stereo geometry. It has been demonstrated that the proposed method is capable of detecting both planar and non-planar road surfaces. Extensive experimental results using three challenging benchmarks (i.e. enpeda, KITTI stereo/flow, and Daimler) show that the proposed road surface detection algorithm outperforms the baseline algorithms both in terms of detection accuracy (up to 23.12%) and runtime performance (up to 95.00%). Next, robust and low complexity algorithm for computing the ego-vehicle’s motion state is proposed. The proposed method estimates the ego-motion of the vehicle by first employing a novel pruning technique to reduce the computational complexity of the corner feature detection process without compromising on the quality of the extracted corner features. A robust and compute-efficient KLT tracker is proposed to facilitate the generation of the feature correspondences. Finally, an early RANSAC termination condition is introduced to the Gaussian-Newton optimization scheme to achieve rapid convergence of the motion estimation process. Evaluations based on the KITTI odometry benchmark show that the proposed visual odometry method outperforms the baseline algorithms both in terms of accuracy (up to 48.36%) and runtime performance. In addition, the proposed algorithm is placed among the top 15% when evaluated using the well-known KITTI odometry platform. Methods for robust and low complexity stereo-vision based obstacle detection and tracking are proposed. Unlike the works that focus only on the detection of vehicles or pedestrians, the proposed obstacle detection method relies on u-v disparity space to detect all obstacles in the scene. A Space of Interest (SOI) is defined to greatly reduce the search space of obstacles prior to employing adaptive hysteresis thresholding and connected component labeling techniques to segment SOI into sets of obstacles. Method for tracking obstacles across frames is also proposed by constructing a distinctive object appearance model. A number of strategies to further increase the distinctiveness and reduce the computational complexity for constructing the object model are also adopted. Finally, an online multi-object tracking framework is proposed by integrating the obstacle detection and data association modules in a robust way. Evaluations using the KITTI tracking benchmark confirm that the proposed obstacle detection and tracking method outperforms the baseline algorithm in terms of tracking accuracy by up to 51.78%. In addition, compared to the baseline algorithm that achieves about 0.23 frame per second (fps), the proposed method lends well for real-time performance with 20 fps. Finally, an efficient and robust risk assessment framework is proposed by integrating the obstacle detection and tracking, and visual odometry methods proposed in this thesis. The Extended Kalman Filter is customized to enhance the robustness of the predicted trajectory of the obstacles for assessing the collision risk. The robustness of collision prediction has been enhanced by accommodating positioning uncertainty. Evaluations based on the KITTI tracking dataset demonstrate that the proposed method are capable of robust and efficient assessment of the collision risk in diverse traffic scenarios. The proposed vision based scene understanding techniques in this research have paved the way towards realizing a real-time capable collision avoidance system that is both affordable and dependable.

Conversational Learning is an intelligent educational assistant application that leverages multimodal AI to personalize the development of communication skills with dynamic real-time interaction experiences. This paper showcases an enhanced version of the system that brings together large language models (LLMs), speech-to-text processing, and contextually aware feedback to deliver immersive, adaptive learning experiences. The system integrates Google Gemini API, Google Text-to-Speech (gTTS) and speech recognition tools to create realistic communication experiences that fit the learner's profile. The system architecture involves a modular approach to present seamless deployment of the system as a web interface using Streamlit and React; while the backend services provide speech analytics and a prompt-driven dialogue. The platform also demonstrated the use of real-time grammar feedback, emotional tone assessment, and quiz customization generated by the AI. Through system validation and usability prior research studies, Conversational Learning has the potential to bridge the gap from knowledge to actionable communication. Most importantly, the system architecture is scalable and privacy-conscious and allows for integration with existing learning systems - an important feature for individual learning at an exciting time for place-based education involving AI.

Depth maps produced by consumer-grade sensors suffer from inaccurate measurements and missing data from either system or scene-specific sources. Data-driven denoising algorithms can mitigate such problems; however, they require vast amounts of ground truth depth data. Recent research has tackled this limitation using self-supervised learning techniques, but it requires multiple RGB-D sensors. Moreover, most existing approaches focus on denoising single isolated depth maps or specific subjects of interest highlighting a need for methods that can effectively denoise depth maps in real-time dynamic environments. This paper extends state-of-the-art approaches for depth-denoising commodity depth devices, proposing SelfReDepth, a self-supervised deep learning technique for depth restoration, via denoising and hole-filling by inpainting of full-depth maps captured with RGB-D sensors. The algorithm targets depth data in video streams, utilizing multiple sequential depth frames coupled with color data to achieve high-quality depth videos with temporal coherence. Finally, SelfReDepth is designed to be compatible with various RGB-D sensors and usable in real-time scenarios as a pre-processing step before applying other depth-dependent algorithms. Our results demonstrate our approach’s real-time performance on real-world datasets shows that it outperforms state-of-the-art methods in denoising and restoration performance at over 30 fps on Commercial Depth Cameras, with potential benefits for augmented and mixed-reality applications.

The Advanced Receiver Autonomous Integrity Monitoring (ARAIM) provides an Aircraft‐Based Augmentation System function for aircraft Global Navigation Satellite System equipment. Currently, the consideration of temporal correlations of test statistics is an important update to the ARAIM baseline algorithm. However, due to the tight budgets of integrity and continuity, the updated ARAIM Fault Detection and Exclusion (FDE) is not satisfactory in availability coverage. The baseline algorithm is based on equal allocation of integrity and continuity, which causes ARAIM to be susceptible to faults and satellite outages. An availability optimisation of ARAIM FDE based on dynamic budget allocation is proposed, aiming to make full use of the prior knowledge provided by the Integrity Support Message (ISM). The contribution of each solution separation test in the ARAIM FDE to the total continuity and integrity risks is estimated, and the continuity and integrity allocation constraints are obtained. The impacts of faults and satellite outages on availability are analysed, and a predictive Protection Level (PL) for ARAIM FDE is proposed for use as an objective function for optimising risk allocation. Then, a risk allocation method based on the multiplier penalty function method is proposed. The performances of the baseline algorithm and the proposed method under the maximum Number of Effective Samples are simulated. For dual‐constellation H‐ARAIM, the optimised Horizontal PL (HPL) is 40% lower than that of the baseline algorithm on average, and the average number of critical satellites supporting RNP 0.3 is less than 1 worldwide. For the three‐constellation V‐ARIAM, the optimised HPL is 27% lower than that of the baseline algorithm on average, and the optimised Vertical PL (VPL) is 18% lower. The proposed method has high availability coverage and can better respond to the requirements for high‐performance navigation services.

Thermal stress plays a prominent role in the reliability of bushings and contributes to the reliability of power transformers, especially during overload conditions. Thus, exploring the temperature distribution in the bushings is essential. This paper proposes a new thermal model to estimate the hotspot temperature (HST) of oil-impregnated paper (OIP) bushings, based on a modified thermal-electrical analogy model. The proposed model is developed based on the finite element method (FEM) to accurately model all fluid flow and internal convection as well as the thermal conduction mechanism. To this end, convection thermal resistances are defined, and their nonlinear characteristics are calculated for different overloading conditions. The proposed approach is applied on a 245 kV, 800 A OIP bushing to analyze not only the normal loading condition but also short-term overloading beyond the rated current. In addition, the oil flow condition with different load currents and transformer top oil temperatures (TTOT) are investigated. The results show a consistent temperature rise with the typical test conditions with no identified problematic situation. However, during overloading, the temperature rise exceeds beyond the permissible limits recommended by IEEE C57.19. Hence, the overloading of transformers may deteriorate the bushing insulation and reduce its lifetime.

Real-world industrial applications impose complex constraints, such as distribution, end-to-end deadlines, and jitter control on real-time systems. Most scheduling algorithms concentrate on single or limited combinations of constraints and requirements only. Offline scheduling resolves complex constraints, but provides only very limited flexibility. Online scheduling on the other hand, supports flexibility, resource reclaiming, and overload handling, but handling constraints such as distribution or end-to-end deadline can be costly, if not intractable. In the paper, we propose a method to efficiently handle soft real-time tasks in offline scheduled real-time systems using a total bandwidth server. In a first step, the offline scheduler resolves complex constraints, reduces their complexity, and provides for guaranteed available bandwidth. The constructed schedule is translated into independent tasks on single nodes with start-times and dead-line constraints only. These are then executed using earliest deadline first, total bandwidth server scheduling at runtime.

A real-time system is overloaded when all the tasks in a workload cannot meet their deadlines, and hence a robust algorithm is essential to maximize the number of tasks that meet their deadlines with the minimum number of miss rates and context switching. Although the Rate Monotonic (RM), Earliest Deadline First (EDF), and Least Laxity First (LLF) algorithms optimally perform and schedule tasks on a non-overloaded system, they have deficient performance when the system is overloaded. Therefore, we propose a new scheduling algorithm for uniprocessor and partitioned multiprocessor systems to address the overload situation. Since the proposed scheduling algorithm operates like EDF non-overloaded conditions, the proposed algorithm is optimal for non-overloaded systems. In addition, the proposed algorithm is robust against overloading situations as it executes the maximum possible tasks in the overload situation instead of missing deadlines of many tasks or burdening context switching to the system. The proposed algorithm allocates a processor to tasks based on the possibility of executing the task. The experimental results demonstrate that the proposed scheduling algorithm maximizes the number of tasks that meet their deadlines in overload conditions without a domino effect and context switching. In addition, the proposed algorithm achieves the lowest miss rate without context switching and the highest efficiency and processor utilization in the overloaded system compared with RM, EDF, and LLF.