原创 Sampling based path planning

The main idea in sampling based path planning methods is to avoid a tedious construction of C-obstacles by sampling

原创 Gazebo使用心得

文章目錄world文件編寫物理屬性設置model文件編寫gazebo中提供的服務 world文件編寫 <?xml version="1.0"?> <sdf version='1.6'> <world name=''>

原创 Online Utility-Optimal Trajectory Design for Time-Varying Ocean Environments

Introduction Task: energy-efficient time-varying current goal on-line Method formulate the general utility-optimal

原创 Recurrent Dropout without Memory Loss

文章目錄Abstract Abstract 目前廣泛採用的regulation方法是 dropout, which is applied to forward connections of feed-forward archite

原创 有限微分

首先介紹一下微分算子: 如果運算作用在函數上,那麼我們就稱這個運算爲算子。 微分算子: df(x)=limh→0f(x+h)−f(x)hdf(x) = \underset{h \rightarrow 0}{lim}\frac{f(

原创 ROS 通信:service

文章目錄Introduction創建service msg創建server node創建client nodeReference Introduction 與基於publisher/subscriber的通信機制topic相比,s

原创 Stochastic Optimal Control (SOC)

文章目錄DefinitionNotelinearly-solvable optimal control (LSOC)Solutionmultiscale or hierarchical Definition The SOC pro

原创 Discrete Planning

No geometric models or differential equations will be needed to characterize the discrete planning problems. State-

原创 [2019 ICRA] Uncertain-Aware Path Planning for Navigation on Road Networks Using Argumented MDPs

文章目錄IntroductionEnvironmentAMDP based planning Introduction 這篇paper的motivation就是planning algorithms忽略position uncer

原创 sklearn 學習筆記 —— Nearest Neighbors

文章目錄IntroUnsupervised Nearest NeighborsNearest Neighbors Classification Intro sklearn 提供了 sklearn.neighbors 這個模塊,這個

原创 clustering 聚類算法

文章目錄clustering 介紹聚類的定義聚類的評價標準Calinski-Harabaz 分數如何確定聚類的個數具體方法Mean-shift(均值遷移)概述Spectral Clustering(譜聚類)概述Hierarchic

原创 ros 命名空間

文章目錄全局命名空間相對名稱私有名稱節點命名空間 全局命名空間 /rosout前面的反斜槓“/”表明該節點名稱屬於全局命名空間。之所以叫做全局名稱因爲它們在任何地方(包括代碼、命令行工具、圖形界面工具等的任何地方)都可以使用。無論

原创 [1996 TRO] Probabilistic Roadmaps for Path Planning in High-Dimensional Configuration Spaces

文章目錄AbstractIntroRelated WorksGeneral methodlearning phasethe construction stepcreating of random configurationsloc

原创 [2002 CDC] Real-Time Motion Planning for Agile Autonomous Vehicles

文章目錄abstractIntro abstract 首先要明確幾個詞在這篇文章中的含義: dynamic system randomized path planning agile We have to take into a

原创 AUV motion planning 筆記 —— MDP

文章目錄Problem statementCurrent field estimationTransition modelWhy do not we use Gaussian distribution?Nominal trajec