Since proportional-integral-derivative (PID) controllers absolutely dominate the controlengineering, numbers of different control structures and theories have been developed to enhance the efficiency of PID controlle...
详细信息
Catastrophic forgetting is a tough challenge when agent attempts to address different tasks sequentially without storing previous information, which gradually hinders the development of continual learning. Except for ...
详细信息
Lane detection separating the roads and positioning each lane is a significant task in Autonomous-driving field. Most of existing methods take lane detection as a semantic segmentation problem, and the methods based o...
Lane detection separating the roads and positioning each lane is a significant task in Autonomous-driving field. Most of existing methods take lane detection as a semantic segmentation problem, and the methods based on encoder-decoder networks achieve robust performances than traditional methods with the disadvantages of large amounts of parameters and high computational cost. To improve efficiency and better adapt to adverse environments, we propose an EDSPnet by compiling the efficient dense module of depthwise dilated separable convolution (EDD module) and dense spatial pyramid (DSP) module. The two proposed modules enable our network to make full use of the contextual information from different feature maps when compared to a chain of layers. Different from symmetrical segmentation networks, we concentrate on the encoder block and a simple decoder is designed to match the input resolution. The decoder is complemented by two methods of upsampling: bilinear interpolation and deconvolution, which guarantee the accuracy and efficiency simultaneously. Our method is evaluated on the TuSimple dataset, and the experimental results show that our network has less parameters and is still efficient even compared with the state-of-the-art semantic segmentation networks.
Vehicle and driver detection in the highway scene has been a research hotspot in the field of object detection in recent years, and it is still a challenging problem in the research of traffic order and road safety. I...
Vehicle and driver detection in the highway scene has been a research hotspot in the field of object detection in recent years, and it is still a challenging problem in the research of traffic order and road safety. In this paper, we propose a novel end-to-end vehicle and driver detection method named VDDNet which is based on Cascade R-CNN and SENet. By introducing FPN structure and SENet attention mechanism in the backbone, the ability of the model to learn effective features is enhanced. It can improve the accuracy of detection in difficult scenes such as weak light, partial occlusion, and low picture resolution. The test results based on the database of highway traffic vehicle and drivers constructed by the Jiangsu Provincial Public Security Department. It shows that the detection method has the AP rate of 91.3% and the Recall rate of 92.4%, which demonstrates the effectiveness of the proposed method in complex highway environments.
Face recognition technology is widely applied in daily life, but in most methods, similarity or affine transformation is employed to align face images according to five facial landmarks. The face alignment module is i...
Face recognition technology is widely applied in daily life, but in most methods, similarity or affine transformation is employed to align face images according to five facial landmarks. The face alignment module is implemented independently, thus it's difficult in end-to-end training. In this paper, the main purpose is to design a towards end-to-end trainable face recognition method based on indoor scenes. Due to that spatial transformer can implement any parametrizable transformation, we joint it with recognition network, making end-to-end training possible. Simultaneously, any prior knowledge on facial landmarks isn't required. The model jointly with spatial transformer can achieve 0.3% higher accuracy than similarity transformation. Most downsampling methods ignore the sampling theorem, making convolutional networks not shift-invariant. We replace max-pooling by MaxBlurPool in spatial transformer network, and the accuracy is improved by 0.25%.
This paper is concerned with the problem of cooperative attitude tracking and vibration reduction for multiple flexible spacecraft without modal variable measurement under a directed communication topology. Firstly, a...
详细信息
This paper is concerned with the problem of cooperative attitude tracking and vibration reduction for multiple flexible spacecraft without modal variable measurement under a directed communication topology. Firstly, a distributed attitude synchronization control law for MRPs representation is proposed in the absence of external disturbances, where the modal information is obtained using a modal variable estimator. To deal with a more practical case that the spacecraft are affected by external disturbances, the control law is further modified by integrating a disturbance observer and a feedforward compensating scheme,in which the lumped disturbances including the external disturbances and errors of the estimated modal information are observed and compensated in finite time. Under this enhanced control scheme, the attitude tracking errors and vibration variables will converge to zero even in the presence of external disturbances, and the controller is continuous and chattering-free. Besides,compared with the existing results, the communication graph in this paper is assumed to be directed and contain a spanning tree with the virtual leader as the root node. A numerical example is illustrated to demonstrate the effectiveness of the proposed results.
In order to adapt to the complex disaster environment, this paper considers the system design of hexapod search and rescue robot. Such hexapod robot is suitable to different kinds of roads and obstacle, which can avoi...
详细信息
In order to enhance the stable operation of the multi-energy complementary microgrid for wind, solar, and diesel storage, reduce operating costs, and solve the problems of large randomness, low accuracy, and slow conv...
In order to enhance the stable operation of the multi-energy complementary microgrid for wind, solar, and diesel storage, reduce operating costs, and solve the problems of large randomness, low accuracy, and slow convergence of traditional microgrid optimization multi-objective decision-making, a differential evolution based on the entropy weight method to determine the weight is proposed. Firstly, we establish a microgrid integrated energy model from the perspectives of system operation stability, economy and environmental protection; combined with the Pareto optimal solution set, multi-objectives are weighted according to the entropy weight method, and the multi-objective optimization problem is transformed into single-objective optimization The problem is to avoid artificial setting of weight factors; the calculation example shows that this method is more economical and reasonable in optimization results, and provides an economic, reliable and environmentally friendly microgrid configuration strategy for users to increase power capacity.
The self-oscillating loop is an important part of the optically pumped cesium magnetometer, and its working characteristics directly determine the accurate measurement of external magnetic field. The design of the sel...
The self-oscillating loop is an important part of the optically pumped cesium magnetometer, and its working characteristics directly determine the accurate measurement of external magnetic field. The design of the self-oscillating loop has been discussed in this paper, including a signal conditioning circuit, a phase shifter and a frequency meter. It can be used to precisely improve the accuracy of resonance signal in a wide range of frequencies from 50 kHz to 350 kHz. The relative error of our system is less than 0.5×10−6 and it has a good prospect in the optically pumped cesium magnetometer.
High-density resistivity meter is a common instrument used in shallow geophysical exploration. At present, the structure of the centralized high-density resistivity meter is bulky, while the distributed instrument nee...
High-density resistivity meter is a common instrument used in shallow geophysical exploration. At present, the structure of the centralized high-density resistivity meter is bulky, while the distributed instrument needs more than 7 core cables, and the long measuring line needs to relay power supply to the intelligent electrode to provide enough voltage, so it is not convenient to use in the field. Aiming at the above problems, a new type of distributed high-density meter is developed, which can realize the power supply, communication, and measurement functions of distributed high-density meter and carry out multi-channel measurement through the time-division multiplexing of a two-wire cable and the short-time power supply of supercapacitor. Finally, the performance of the instrument was verified by experiments.
暂无评论