Cognitive Robotics最新文献

Revenue loss is a major issue for any country. Conversion of this loss into utilization would prove to be a huge benefit to the country. In view of this fact, the economic load dispatch problem draws much attention. Substantial reduction in fuel cost could be obtained by the application of modern heuristic optimization techniques for scheduling of the committed generator units. In this study, two cases are taken named three-unit system and six-unit system. The fuel cost for both systems compared using conventional lambda-iteration method and PSO method. These calculations are done for without transmission loss as well as with transmission losses. In the end, the fuel cost for both methods compared to analyze the better one from them. All the analyses are executed in MATLAB environment.

Deep learning architectures have yielded a significant leap in target detection performance. However, the high cost of deep learning impedes real-world applications, especially for UAV and UGV platforms. Moreover, detecting small targets is still of lower accuracy in contrast to the large ones. Aiming to comprehensively handle these two issues, a novel SP-CBAM-YOLOv5 architecture is proposed. The main novelty of our hybrid model lies in the cooperation of the attention mechanism and the typical YOLOv5 architecture, which can largely improve the performance of the small target detection. Moreover, the depth convolution and knowledge distillation are jointly introduced for lightening the model architecture. To evaluate the performance of our proposed SP-CBAM-YOLOv5, we built a novel dataset containing challenging scenes of power engineering. Experimental results on this benchmark demonstrate that our proposed SP-CBAM-YOLOv5 achieves a competitive performance in contrast to the other YOLO architectures. Besides, our lightweight YOLOv5 has more than 70% decrease of parameters. Moreover, the ablation study is conducted to demonstrate the compact architecture of SP-CBAM-YOLOv5.

In the process of multi-UAVs cooperative reconnaissance operations, due to the limited battery capacity and computing resources of the unmanned aerial vehicle (UAV), processing tasks can not only lead to excessive delay, but also increase the energy consumption of the UAV, which reduces the endurance time of the UAV. Therefore, we have proposed a mobile edge computing (MEC) system architecture composed of single unmanned helicopter (UH) and multiple reconnaissance UAVs. Among them, the UH as a MEC server to provide computing services for reconnaissance UAVs. By solving the computing offloading strategy problem of multi-UAVs, the objective is to minimize the weighted sum of energy consumption and delay for the multi-UAVs' task execution. In solving the problem, previous heuristic algorithms such as the Particle Swarm Optimization (PSO) are often used as basic algorithms for research, but they tend to converge early, fall into local optimum easily, and have low solution accuracy, making it difficult to obtain the optimal offloading strategy. Therefore, this paper proposes an improved bat algorithm (IBA) with fast convergence ability and global search ability. Through the simulation experiments and comparative analysis of PSO, BA, IPSO and IBA, it is proved that the IBA is more accurate, stable, and efficient in solving this problem based on the system architecture proposed in this paper, and effectively reduces the weighted sum of energy consumption and delay for the multi-UAVs' task execution.

A critical problem for fault diagnosis is caused by the feature shift under different working conditions, which significantly degenerates the diagnosis accuracy in practice. Aiming to solve this problem, this paper proposes a novel Transfser Learning (TL) framework with Dynamic Multiscale Representation (DMR) for fault diagnosis. This model draws the inspiration from the shared learning and transfer learning, processing information captured and exploited by multiscale signal factors. In particular, a novel multi-path merging network is proposed to generate dynamic weights for fusing multiscale factors. To drive this generation, and to control the extent of the shared fusion, the Multi-gate Mixture-of-Experts (MMoE) is introduced to model the tradeoff between scale-specific representation and inter-scale correlation. A transfer learning backend is also introduced to align cross-domain features, which enables proposed method to diagnose faults across distinct working conditions. Experiments evaluate the fault-diagnosis performance. Our primary, ablation and interpretation evaluations comprehensively indicate the robustness and flexibility of the proposed method to diverse fault diagnosis applications. Especially, the proposed method achieves 4.71% and 3.86% improved to the second best one (MSSLN) on the PHM2009 and MCP datasets, respectively.

Mental simulation of actions is a powerful tool for allowing cognitive agents to develop Prospection Capabilities that are crucial for learning and memorizing key aspects in challenging actions. In particular, this study focuses on the initial or final posture of actions and provides a computational tool that allows an agent to evaluate their feasibility and appropriateness. Such tool is a kinematic network, equivalent to an internal body schema, that allows a cognitive agent to generate simulation-states that reach the goal with a comfortable final posture, by exploiting the redundancy of the kinematic network. This is obtained by activating and integrating in the network dynamics three types of virtual force fields: 1) Focal force field applied to the end-effector, related to the goal of the action; 2) Range of Motion force fields, applied separately and independently to each degree of freedom in order to preserve the natural joint limits; 3) Postural force field, applied to the pelvis area, for maintaining the projection of the center of mass of the body model inside the support base. The efficacy of this approach is demonstrated in relation to a simple task: reaching a heavy load in order to lift it and then shifting it forward before dropping it on a table. The mental simulation model attempts to provide a kinematic template compatible with the overall plan and the postural/articular constraints, as a function of the initial position of the body relative to the load. The simulation may fail and this indicates that the chosen initial posture is inappropriate for the task. Successful simulations can also be evaluated in terms of precision and effort by monitoring the peak torque required of each joint actuator. Optimal or at least sub-optimal solutions can be memorized in episodic memory, thus accruing the know-how of the agent.

Artificial Intelligence (AI), Machine Learning (ML), and Deep Learning (DL) have revolutionized the field of advanced robotics in recent years. AI, ML, and DL are transforming the field of advanced robotics, making robots more intelligent, efficient, and adaptable to complex tasks and environments. Some of the applications of AI, ML, and DL in advanced robotics include autonomous navigation, object recognition and manipulation, natural language processing, and predictive maintenance. These technologies are also being used in the development of collaborative robots (cobots) that can work alongside humans and adapt to changing environments and tasks. The AI, ML, and DL can be used in advanced transportation systems in order to provide safety, efficiency, and convenience to the passengers and transportation companies . Also, the AI, ML, and DL are playing a critical role in the advancement of manufacturing assembly robots, enabling them to work more efficiently, safely, and intelligently. Furthermore, they have a wide range of applications in aviation management, helping airlines to improve efficiency, reduce costs, and improve customer satisfaction. Moreover, the AI, ML, and DL can help taxi companies in order to provide better, more efficient, and safer services to customers. The research presents an overview of current developments in AI, ML, and DL in advanced robotics systems and discusses various applications of the systems in robot modification. Further research works regarding the applications of AI, ML, and DL in advanced robotics systems are also suggested in order to fill the gaps between the existing studies and published papers. By reviewing the applications of AI, ML, and DL in advanced robotics systems, it is possible to investigate and modify the performances of advanced robots in various applications in order to enhance productivity in advanced robotic industries.

Digital Twin is a virtual representation of objects, processes, and systems that exist in real-time. While Digital Twin can represent digital objects, they are often used to connect the physical and digital worlds. This technology plays a vital role in fulfilling various requirements of Industry 4.0. It gives a digital image of a factory's operations, a communications network's activities, or the movement of items through a logistics system. This paper studies Digital Twin and its need in Industry 4.0. Then the process and supportive features of Digital Twin for Industry 4.0 are diagrammatically discussed, and finally, the major applications of Digital Twin for Industry 4.0 are identified. Digital Twin sophistication depends on the process or product represented and the data available. Manufacturers can learn how assets will behave in real-time, in the physical world, by putting sensors on particular assets, gathering data, creating digital duplicates, and employing machine intelligence. They can confidently make wise judgments, which helps improve company performance. Digital Twin assesses material usage to save costs, discover inefficiencies, replicate tool tracking systems, and do other things. Manufacturers construct a digital clone for specific equipment and tools, exclusive products or systems, entire procedures, or anything else they want to improve on the factory floor. Sensors and other equipment that collect real-time data on the state of the process or product collect this information, which on the other hand, must be handled and processed appropriately. It is made feasible by IoT sensors, which collect data from the physical environment and transmit it to be virtually recreated. This information comprises design and engineering details that explain the asset's shape, materials, components, and behaviour or performance.

Scientific visualization is important in modern technological activities and engineering exploration. Due to the dark and high-pressure characteristics of deep sea, it is difficult to visualize the entire operation of deep-sea equipment. Thus, it is of great necessity to use virtual simulation technology to help people understand the operation process of some deep-sea exploration equipment on the sea floor. In this paper, science, art, and new media are combined through artistic rendering, visual processing, and the technology of virtual reality (VR) and holography, which makes the exploration of the latest deep-sea lander and intelligent submersible named “Luling” look more intuitive and smart and have more visual impact and expression. Apart from that, automatic manipulation videos of the rover robot in the deep sea captured by the Luling are effectively nested to realize the goal of virtual and real presentation. The designed scientific visualization of deep-sea equipment can not only adapt to the display output of VR, mobile phones, TV, 360° showcase, and other platforms, but also achieve immersive experience and virtual simulation learning through HTC Vive VR equipment. The technology and design way of scientific visualization in this paper is universal and suitable to the same kind of engineering simulation.

In realm of researches involved in autonomous tractor-trailer robot, novel purpose of this research has been dedicated to motion resistance force of rubber tracked undercarriage of the robot. Hence, the motion resistance force was ascertained as affected by operational variables of robot forward speed (0.17, 0.33 and 0.5 m/s) and payload weight (1, 2, 3, 4 and 5 kN). Analytical results clarified that meaningful contribution of payload weight to the motion resistance force (15.26–28.05 N) was marginal (< 8 times) in comparison with that of robot forward speed. Hence, adjustment of the forward speed than payload weight is suggested as priority. Modeling results described that combinatorial effect of robot forward speed and payload weight on the motion resistance force was synergetic. This disclosed linear increasing dependency of the motion resistance force on concurrent proliferation of robot forward speed and payload weight. Overall, these results are profitable for redesign and performance optimization of tractor-trailer robot with rubber tracked undercarriage in order to proliferate autonomous transportation capacity of payloads, especially for indoor and outdoor shipping and warehouse of factories and industrial environments.

Over the past decade, the utilization of mobile robots in commercial and defense industries has rapidly increased. These robots are purpose-built to perform specific tasks and have proven to be particularly valuable in dangerous environments where human presence may be problematic. However, identifying hazardous areas for workers, soldiers, and emergen- cies during disasters and providing real-time surveillance data remain significant challenges. Conventional approaches, such as manual surveillance and mapping uncharted territories are time-consuming and susceptible to human error. UGVs enable standoff operations, which lowers or eliminates these problems in demanding, and hazardous conditions. This paper discusses the design and development of Rakshak: a modular UGV as a first response mechanism for 360° of real-time surveillance by mapping unknown areas and small- payload-based logistics operations. Teleoperation of the UGV is via radio transmission, a reliable and efficient method of communication. The modular design of the UGV allows for flexibility in adapting to various applications. Data acquisition and transfer to the mobile application are accomplished through Wi-Fi communication.