人工智能研究院“文一研究生論壇”學術係列講座第1期
上傳時間:2020-10-23 瀏覽次數:10

報告題目:A Potential Game based Distributed Task Scheduling    Model of Multi-Earth Observing Satellites

報告人:馮蕊

報告摘要:We consider that multi-earth observing satellites (EOSs) follow their orbits to observe a set of targets on the earth to complete the observing tasks. The problem is studied of achieving the maximal task observing ratio with least overlapped target for all EOSs. To describe the targets observing sequence of each EOS, we establish a individual task graph model according to the EOSs' configuration. In addition, based on the proposed task graph model, we design the global and local utility function, respectively. Based on the game theory, it is proved that each EOS's utility and the global utility satisfied the relationship of an exact potential game. To improve the target observing ratio, a distributed task scheduling model is proposed to find each agent's local optimal solution, which converges to the global optimum.


報告題目:Epileptic State Classification by Fusing Hand-crafted and Deep Learning EEG Features

報告人:胡丁寒

報告摘要:Seizure onset detection and epileptic preictal prediction based on electroencephalogram (EEG) signals have been a challenge problem in the research community. We study a novel epileptic states classification algorithm based on the multichannel EEGs representation using multiple hand-crafted features. The frequency domain features (MAS and MPSD) and timescale features (WPFs) are combined together for multichannel EEG representation. Multiple diverse pre-trained DNNs have been adopted for feature transfer learning on the fused EEG image feature and a new hierarchical neural network has been developed for discriminative feature learning and epileptic state classification.


報告題目:Multi-modal Physiological Signals based Fear of Heights Analysis in Virtual Reality Scenes

報告人:鄭潤澤

報告摘要:Fear of heights (FoH) analysis and its association to physiological signals can better help understand people’s emotion and quantify human’s behaviour, which have been found important in many applications, such as disease analysis, affective computing, etc. Existing studies are mainly on how to alleviate FoH while little literature on FoH analysis has been reported in the past. In this paper, we present the studies of correlation on FoH to multi-modal physiological signals in the virtual and reality (VR) scenes. To stimulate the FoH to participants, 4 types of VR scenarioses that consist of the virtual scene of the VR game “Richie’s plank experience” and the realistic stimulus of hitting by basketball are adopted in the experiment. The synchronized eye movement (EMO), pupil, and electrocardiogram (ECG) of 17 healthy subjects with an even mix of men and women are recorded for FoH analysis. The multi-modal physiological signals based analysis reveals that: 1) the physiological features, including the pupil diameter, the power spectral densities (PSD) of EMO, pupil, ECG, the mean of EMO, etc., have obvious changes in different VR scenarioses (ground/high-altitude scenes), 2) machine learning models learning on multi-modal physiological signals combining with feature optimization can achieve.


報告題目:一種工業機器人關節軌跡規劃方法

報告人:鄔尚良

報告摘要:運動規劃是工業機器人領域的重要技術,規劃結果對機器人性能有著深刻的影響。將任務分解成若幹個子任務,並在完成對任務的調度排序後,針對任務間的過渡過程進行關節運動規劃。首先,采用RRT算法進行關節變量的路徑規劃,難點在於機械臂連杆的碰撞檢測。將連杆和障礙物分別用橢球近似,並建立連杆橢球關於關節變量的運動方程,以及機械臂避障的數學模型。該方法的優勢在於碰撞檢測的計算負荷小。然後,麵向關節軌跡規劃,采用一種改進的樣條參數化形式進行軌跡擬合,以實現軌跡的指標優化,並減少求解的計算負荷。


報告時間20201030日(星期五)13:30-15:30

報告地點:杭州電子科技大學文一人工智能研究院一樓報告廳


科學研究
聯係我們