Deep brain stimulation (DBS) is a promising therapy for treatment-resistant major depressive disorder (MDD). MDD involves the dysfunction of network that can exhibit complex nonlinear neural dynamics in multiple frequency bands. However, current open-loop and responsive DBS methods cannot track multiband MDD, leading to imprecise regulation symptoms, variable treatment effects among patients, high battery power consumption.Here, we develop closed-loop brain-computer interface (BCI) system...

The dynamic response of a multilayer finite plate resting on two-parameter viscoelastic foundation to moving variable loads is obtained analytically/numerically. governing differential equation multi-layer derived firstly. Then, the analytically using mode superposition and Duhamel's formula, followed by solving numerical method. Next, proposed solution verified comparing degraded with both field measurement data known theoretical results. Three different loading models are assessed select...

Abstract In this paper, the control problem of multi‐robot systems under temporal logic tasks with limited time and constraints is studied, where each robot required to reach specified region in a given avoid collision all time. Since cooperative avoidance task one depends on other robots' behaviors, satisfaction may be conflicting. work, distributed model predictive (DMPC) strategy proposed for conflicting tasks. First, signal (STL) adopted formally describe Based robust semantics STL...

Objective. Neuromodulation systems that use closed-loop brain stimulation to control states can provide new therapies for disorders. To date, has largely used linear time-invariant controllers. However, nonlinear time-varying network dynamics and external disturbances appear during real-time stimulation, collectively leading model uncertainty. Real-time uncertainty degrade the performance or even cause instability of Three problems need be resolved enable accurate stable under First, an...

In this work, we present a multi-sensor fusion based localization framework for robots in both indoor and outdoor environment. This work aims to utilize the advantages of LiDAR, GNSS IMU sensors order achieve best state estimation varied environments. The proposed frame is composed two parts: feature-based LiDAR simultaneous mapping (SLAM) filter-based estimation. We first establish priori point cloud map on SLAM, ensure consistency coordinate system by adding constraints back-end...

This paper investigates the problem of connectivity-preserving flocking multiple autonomous agents with second-order dynamics. First, inverse power iteration algorithm is formulated in a completely distributed manner to estimate algebraic connectivity, i.e., second smallest eigenvalue group Laplacian, as well corresponding eigenvector. Furthermore, gradient-based algorithms that exploit decentralized eigenvalue/eigenvector estimation are developed both steer agent desired motion and maintain...

This paper aims to propose an online 3D mesh reconstruction algorithm based on lidar sensors for large-scale scenes. In comparison point clouds, meshes provide stable geometry and richer semantic information. To address the issues related sparse inaccurate normal estimation, slow calculation, we introduce a new system. Our approach combines robot's motion with method, enabling us generate high-quality from cloud in frame-by-frame manner then stitch them together form global mesh. evaluate...

To evaluate the mechanical behavior and load transfer mechanisms of a geosynthetic-reinforced pile-supported (GRPS) embankment under traffic load, an analytical model is developed based on elastodynamic theory, its solution derived rigorously. Since present continuum based, displacement stress distributions in can be obtained analytically. Additionally, takes anisotropic property reinforced layer into consideration. The governing equations are solved by introducing Fourier expansions field...

Developing closed-loop brain stimulation systems can benefit the treatment of neurological and neuropsychiatric disorders facilitate functions. Current designs controllers have used time-invariant linear models activity to devise non-adaptive controllers. However, unmodeled nonlinear dynamics happen during real-time control, leading uncertainty in model. cannot track model are not robust noise, both which compromise their control performance. Here, within an ℒ <inf...

Abstract Objective . Closed-loop deep brain stimulation (DBS) is a promising therapy for Parkinson’s disease (PD) that works by adjusting DBS patterns in real time from the guidance of feedback neural activity. Current closed-loop mainly uses threshold-crossing on-off controllers or linear time-invariant (LTI) to regulate basal ganglia (BG) Parkinsonian beta band oscillation power. However, critical cortex-BG-thalamus network dynamics underlying PD are nonlinear, non-stationary, and noisy,...

Abstract Objective.&amp;#xD;Developing an efficient and generalizable method for inter-subject emotion recognition from neural signals is emerging challenging problem in affective computing. In particular, human subjects usually have heterogeneous signal characteristics variable emotional activities that challenge the existing algorithms achieving high accuracy. &amp;#xD;Approach. &amp;#xD;In this work, we propose a model-agnostic meta-learning algorithm to learn adaptable...

This paper proposes a novel control scheme with focus on the uncertainty of complex robot system, which combines cascaded cerebellar model articulation controller (CMAC) variable structure (VSC). Firstly, an improved CMAC is used to learn and it as feed-forward compensator, fast tracking error convergence better learning stability obtained through use CMAC. Then, VSC term reduce effect estimate unrepeatable disturbances. The law chosen based Lyapunov direct method. Simulation 6-6 parallel...

Information value model is a widely used landslide susceptibility assessment method. However, it neglects the weights of different factors. This paper proposes weighted information based on AHP. The impact seismic intensity (SC), land-use types (LUT), lithology types(LT), fault density (FD), terrain (T) and rainfall(R) in 512 Wenchuan earthquake-hit 10 degree region Longmenshan zone(LFZ) assessed with GIS spatial analysis. According to these controlling factors, study area can be divided...

Abstract Closed-loop deep brain stimulation (DBS) is a promising therapy for Parkinson’s disease (PD) that works by adjusting DBS patterns in real time from the guidance of feedback neural activity. Current closed-loop mainly uses threshold-crossing on-off controllers or linear time-invariant (LTI) to regulate basal ganglia (BG) beta band oscillation power. However, critical cortex-BG-thalamus network dynamics underlying PD are nonlinear, non-stationary, and noisy, hindering accurate robust...