This paper presents a light-weight convolutional neural network for SEM imaging. Combining nanorobotic manipulation systems with the scanning electron microscope makes precise micro-nano manipulation based on visual feedback become possible. However, the contradiction between imaging speed and resolution hinders the real-time of micro-nano manipulation, which further obstacles to the realization of applying it in mass manufacturing situations. In this paper, we propose a super resolution method called SRSEMNet1 to enhance the imaging quality of SEM by super-resolving low resolution SEM images. In this way, high-quality SEM images could be obtained while consumes less time than acquiring them straightly from SEM, which contributes to the real time of micro-nano manipulation. We analyze the residual learning, skip connections, CBAM, and different loss functions to optimize the network. The method was tested in both well accepted benchmark dataset Urban100 and our own made dataset SSRID for SEM imaging.
Based on research of error control technology,In this thesis,we implement data packet's coding and decoding based on Reed-Solomon code through looking chart;and emulate the real-time error control systems that use error control methods of FEC and FEC/ARQ. Experiments results prove that coding and decoding based on Reed-Solomon code through looking chart can improve the efficiency of coding and decoding,and experiments results also validate FEC and FEC/ARQ error control methods, that are based on the Reed-Solomon code, have fine characteristics and fine real-time, and decrease the code error ratio 5% at least,develop the data reliability and satisfy the QoS constraints.
Bladder cancer is characterized by high rates of recurrence and multifocality. Immunogenic cell death (ICD) of cancer cells has emerged as a promising strategy to improve the immunogenicity of tumor cells for enhanced cancer immunotherapy. Although photosensitizer-based photodynamic therapy (PDT) has been validated as capable of inducing ICD in cancer cells, the photosensitizers with a sufficient ICD induction ability are still rare, and there have been few reports on the development of advanced photosensitizers to strongly evoke the ICD of bladder cancer cells for eliciting potent antitumor immune responses and eradicating bladder carcinoma in situ. In this work, we have synthesized a new kind of endoplasmic reticulum (ER)-targeting aggregation-induced emission (AIE) photosensitizer (named DPASCP-Tos), which could effectively anchor to the cellular ER and trigger focused reactive oxygen species (ROS) production within the ER, thereby boosting ICD in bladder cancer cells. Furthermore, we have demonstrated that bladder cancer cells killed by ER-targeted PDT could serve as a therapeutic cancer vaccine to elicit a strong antitumor immunity. Prophylactic vaccination of the bladder cancer cells killed by DPASCP-Tos under light irradiation promoted the maturation of dendritic cells (DCs) and the expansion of tumor antigen-specific CD8
Pavement structure bearing capacity is an important evaluation parameter in pavement design, construction, maintenance management, and reconstruction, and is generally expressed by the pavement deflection value. Some of the current road bearing capacity detection equipment have high detection accuracy, but the detection speed is slow, they cannot achieve real-time continuous detection; and some detection speeds are fast, but the measurement accuracy is easily affected by the pavement roughness and vehicle vibration. Moreover, the detection result is the pavement displacement, which cannot directly reflect the comprehensive modulus of the pavement structure. In this paper, firstly, a two-stage jump mechanical model of “machine-pavement” system is established in order to develop a device that can simulate the real driving load and continuously test the bearing capacity of pavement structure, and the main factors affecting the acceleration response of vibrating drums were determined through analysis. Then, a finite element model of the “machine-pavement” system was established to overcome the difficulty in obtaining the parameters such as vibrating weight, equivalent stiffness, and equivalent damping of pavement structure in numerical solution of dynamic model. Next, the mean value A of the maximum acceleration signal of the vibrating drum and the coefficient of variation acv of the maximum acceleration signal were selected as evaluation indicators to analyze the change trend of the maximum acceleration of the vibrating drum with the excitation frequency and excitation force under different composite modulus of pavement structures. Finally, the relationship between the composite modulus E of the pavement structure and the maximum acceleration A of the vibrating drum was obtained by simulating the pavement structure with the composite modulus ranging from 100 MPa to 2900 MPa, and the accuracy of this relationship was verified by field tests. The research showed that the acceleration signal of the vibrating drum had a good fitting relationship with the bearing capacity of the pavement structure when the testing device with the vibrating drum mass of 100 kg, the exciting frequency of 60 Hz, and the exciting force of 650 N jumped on the pavement structure, and the error was about 20% after comparing with the results of Benkelman beam testing, which basically met the engineering requirements. Therefore, the device can be used to continuously detect the bearing capacity of pavement structures.
Reinforcement Learning (RL) has been applied to robotic arm control, which enables the agent to learn an effective policy to solve complex tasks. However, it requires constant interaction with the environment leading to low sample efficiency. In this paper, we propose a robotic arm control approach based on planning via lookahead search, which is a model-based RL algorithm to improve the sample efficiency. The approach builds an environment model in order to obtain the dynamics of the environment. Thus the model can be used to plan future actions by a tree-based search. The experiments show that our approach can solve the task of robotic arm control with less environmental samples.
This paper reports on the design and fabrication of buckled metal strain gauges formed using a thermally contracting shape memory polymer (SMP) via inkjet printed silver nanoparticles to create sensors realized in a large initial compressive stress state. Printed devices are placed into an oven to simultaneously sinter the nanoparticles and shrink the substrate into its final dimensions creating devices with feature sizes unobtainable by direct printing. An unconventional offset serpentine strain gauge geometry is presented which overcomes the local anisotropy of SMP contraction. A prototype process is then employed to realize and demonstrate functional strain gauges using printed silver on SMP. The simple fabrication process allows for the realization of metal strain gauges pre-compressed through a substrate size contraction of 2.5 for the first time.
The Mixed Programming for ControlBuild(CB) and Matlab can take advantage of the efficiency in simulation and development.This paper focuses on discussing mixed programming between CB and Matlab.Therefore,the related protocols,implementation steps,pivotal problems in the course of programming and the corresponding strategies are presented.Finally,the advantages and disadvantages of them and their application are proposed.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)