This letter proposes a dynamic system approach to learn point-to-point motions while keeping the stability of the dynamic system. The proposed approach is grounded on a Learning from Demonstration (LfD) method based on a neural network, which gets a better reproduction performance while guaranteeing the generalization ability. The proposed approach has been experimentally validated on the LASA dataset and by the "pick-and-place" task of Franka Emika robot, and experimental results demonstrate that: (1) compared with the state-of-the-art results, the trajectory generated by the proposed approach achieves higher accuracy (approximately 24.79%) in terms of the similarity with respect to the demonstration; (2) the proposed approach can handle high dimensional data and learn from one or more demonstrations; (3) the proposed approach can guarantee the performance regardless of the variation of starting points even in the case of high dimensional complex motions.
Various cognitive systems have been designed to model the position and stiffness profiles of human behavior and then to drive robots by mimicking the human's behavior to accomplish physical human–robot interaction tasks through a properly designed impedance controller. However, some studies have shown that variable stiffness parameters of the impedance controller can cause the violation of the passivity constraint of the robot states, and make the robot's stored energy exceed the external energy injected from the human user, thus leading to the unsafe human–robot interaction. To solve this problem, this article proposes a novel passive model-predictive impedance control method including two control loops. In the bottom-loop of the proposed controller, the robot is driven by a variable impedance controller to achieve the desired compliant interaction behavior. In the top-loop of the proposed controller, the model-predictive control (MPC) is used to ensure that the robot states satisfy the passivity constraint by calculating a complementary torque to limit the stored energy of the robot. The passivity of the closed-loop robot system and the feasibility of MPC are guaranteed by theoretical analysis, ensuring the safety of the robotic movement in the human–robot interaction. The effectiveness of the proposed method is demonstrated by the simulation and experiment on the Franka Emika Panda robot.
In this paper, we study the group consensus for networked Euler-Lagrange systems under a general nonsymmetric directed graph. The relationships between the interactive agents in the same group are cooperation while the relationships between the interactive agents among different group can be either cooperation or competition. Two assumptions are presented to achieve group consensus, respectively, interaction topology among groups being acyclic and the associated directed graph satisfying the in-degree balance condition. We propose a distributed adaptive control algorithm for each agent without relative velocity information. Simulation results are provided to demonstrate the effectiveness of the proposed control algorithms.
This paper mainly investigates the group consensus of Euler-Lagrange multi-agent systems based on the dynamical event-triggered controller. All agents in the system are divided into several groups. The agents in the same group converge to one state and those in the different groups may not coincide. To use fewer communication resources, an event-triggered controller using the dynamical event-triggered law is designed. Several mild assumptions are made to achieve the group consensus. And Lyapunov analysis is given to prove that the group consensus can be achieved under the designed control algorithm. Compared to existing results, this paper mainly designs the dynamical event-triggered law to achieve the group consensus rather than consensus or tracking, and the concerned agent is the nonlinear Euler-Lagrange system.
Introduction Liquid biopsy research contributes to precision medicine initiatives and enables minimally invasive and inexpensive sampling compared to traditional tissue biopsy. However, the low amount of circulating tumour nucleic acid fragments in the blood presents significant challenges for accurate variant detection using NGS technology. Utilisation of both cell free (cf) DNA and cf RNA requires methods capable of interrogating both types of analytes to maximise the utility of each blood sample. We described here a two days sample-to-report workflow using Oncomine Pan-Cancer Cell-Free Assay that surveys oncology variants across multiple tumour types and simultaneously detects single nucleotide variants (SNVs) and structural variants such as copy number variations (CNVs), gene fusions as well as exon skipping. Material and methods Cell Free Total Nucleic Acid (cfTNA) was extracted using MagMAX Cell-Free TNA Isolation Kit. Internal 0.1% or 0.5% cfDNA reference materials were used to evaluate SNV sensitivity and specificity. For CNV sensitivity and specificity evaluation, cfDNA from CNV positive cell lines were titrated into normal donor plasma cfTNA background. All controls were verified with orthogonal assays of dPCR. Libraries were manually prepared and sequenced with Ion Chef and S5 XL System. Data analysis was performed in Torrent SuiteTM 5.6 and Ion Reporter 5.6. Results and discussions The Pan-Cancer Cell Free Research Assay utilised a single-pool multiplex assay to query more than 900 tumour driver and resistance hotspots. The broad content panel encompasses SNVs, CNVs, fusions, exon skipping as well as expanded coverage of TP53 exon regions for TP53 mutation analysis. The entire work flow (sample-to-report) could be as less as 30 hours and is compatible with Oncomine knowledgebase that allows customers easy access to the variant reports. The limit of detection (LOD) of 0.5% and 0.1% allelic frequency can be achieved with sensitivity of >99% and 90% respectively, both specificity >99.9%. Detection as low as 1.3 fold gene amplification can be obtained for 12 CNV targets with sensitivity of >96% and specificity of >99%. For fusion detection, as low as 1% RNA fusions in cfTNA can be achieved. Conclusion The Pan-Cancer Cell Free Research Assay provides an easy and quick NGS workflow that simultaneously analyses SNVs, CNVs, gene fusions and exon skipping across 52 genes associated multiple cancers. The 0.1% LOD enables accurate detection of low-abundance tumour variants for liquid biopsy research.
When traditional mathematical statistic method is used to build fertilization model, the structure design and factor choosing largely rely on prior knowledge of experts in the field. Consequently, the result is somewhat casual and subjective. It is very difficult to settle this universal nonlinear and uncertain problems in fertilization. A new network for fuzzy-neural system which is easy to distill the fuzzy rules is proposed. The network structure is adjusted by FBP(Fuzzy Back Propagation) learning algorithm to acquire network parameters and variable weights of the initial fertilization model. For an optimal fertilization model comes into being, IIP(Improved iterative pruning) algorithm which is applied can lessen the network structure and reduce the complexity of compute to speed up the respond rate of output. Soybean experiments in different plants and years show that this approach can build very accurate model without any prior knowledge, which contributes as theoretical in countryside.
'/%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)
