[{"data":1,"prerenderedAt":117},["ShallowReactive",2],{"method-msckf2_2013":3},{"method":4,"reference":59,"equipment":79,"figures":116,"results":48},{"id":5,"label":6,"shortName":7,"title":8,"year":9,"era":10,"cluster":11,"scope":12,"keyIdeaZh":13,"keyIdeaEn":14,"fulltextStatus":15,"publicationStatus":16,"recommendation":17,"constructionRelevance":18,"validationEnvironment":19,"strengths":22,"limitations":27,"sensors":32,"platform":35,"estimator":38,"association":39,"timeModel":40,"deskew":41,"loopClosure":42,"globalOptimization":43,"mapRepresentation":44,"prior":45,"outputGeometry":46,"compute":47,"codeUrl":48,"codeLicense":49,"relatedVersions":50},"msckf2_2013","Li & Mourikis, 2013","MSCKF 2.0","High-precision, consistent EKF-based visual-inertial odometry",2013,"classic","C08","odometry","論文比較兩類以 EKF 為基礎的視覺慣性里程計（VIO）：狀態含特徵點的 EKF-SLAM，以及只保留滑動視窗位姿的 MSCKF，並以蒙地卡羅模擬顯示 MSCKF 在精度、一致性與運算量上都較佳。作者證明兩者的線性化模型都讓繞重力軸的偏航角看似可觀測，濾波器因此低估不確定度。MSCKF 2.0 以閉式 IMU 誤差狀態轉移矩陣，並在計算 Jacobian 時固定使用位置與速度的首次估計值，恢復正確的不可觀測子空間；同時把相機與 IMU 外參放入狀態線上估計。","MSCKF 2.0 restores the correct 4-DOF unobservable subspace of EKF-based VIO by evaluating Jacobians at first (propagated) estimates with a closed-form IMU transition matrix, and estimates camera-IMU extrinsics online, giving a consistent real-time monocular VIO that beat EKF-SLAM variants and a fixed-lag smoother.","full_text_reviewed","peer_reviewed_published","background","論文未報告營建工地或室內測試；真實實驗為車頂相機與 IMU 在市區行駛 21.5 km，並以 GPS-INS 為參考，其餘結果為依真實資料建構的蒙地卡羅模擬。其價值在於說明濾波式 VIO 的一致性與相機 IMU 外參線上估計，對現場以 VIO 提供位姿先驗時如何解讀共變異數有參考意義（推論）。",[20,21],"simulation","independent_reference",[23,24,25,26],"In the Cheddar Gorge simulation MSCKF 2.0 reached 97.7 m position RMSE and NEES 6.53 versus 133.4 m and 50.97 for an iterative fixed-lag smoother that used about five times more computation (Table 3; Sec. 8.2)","Performance nearly identical to an 'ideal' MSCKF that uses true states for Jacobians (Table 2; Sec. 8.1)","Online camera-IMU calibration gave pose accuracy almost equal to perfectly known calibration (Table 4; Sec. 8.3)","Real 21.5 km drive: largest position error about 58 m (0.28% of distance) versus about 230 m (MSCKF) and 202 m (FLS) (Sec. 9)",[28,29,30,31],"Odometry only: loop closure and re-observation of revisited areas are not handled (Sec. 3.2)","Position and yaw remain unobservable, so their uncertainty (and drift) grows over time even for MSCKF 2.0 (Sec. 8.2)","Discrete-time IMU integration needs assumptions about signal behaviour between samples, so some approximation is unavoidable (Sec. 5.2)","Evaluated on vehicle motion only; no handheld, indoor or aerial tests (inference from Secs. 3.4, 8, 9)",[33,34],"monocular camera","IMU",[36,37],"vehicle (camera and IMU on a car roof; 21.5 km urban drive in Riverside, CA)","simulation built from real vehicle datasets (5.5 km urban dataset; 29 km Cheddar Gorge dataset)","multi-state-constraint EKF: IMU state plus a sliding window of past IMU poses; feature errors removed by left-nullspace projection; Jacobians evaluated at first (propagated) estimates of position and velocity so the linearised model keeps a 4-DOF unobservable subspace; closed-form IMU error-state transition matrix; camera-to-IMU rotation and translation in the state (Secs. 3.3, 5, 7)","Shi-Tomasi corners matched by normalised cross-correlation in the real experiment (about 290 features per image); each feature used once its track ends, triangulated by Gauss-Newton; Mahalanobis gating at the 95th percentile of chi-square with 2N-3 degrees of freedom (Secs. 3.3, 9)","discrete poses at image times; IMU propagation assumes linearly varying signals between samples, orientation by 4th-order Runge-Kutta and position and velocity by Simpson integration (Sec. 5.2)","not_applicable","none (VIO; loop closure is left to a separate algorithm, Sec. 3.2)","none","none; features are never kept in the state vector, only a sliding window of poses","none; manually measured camera-IMU extrinsics used only as initial values for online estimation (Sec. 9)","IMU pose, velocity, biases and camera-IMU extrinsics with covariance; no map or point cloud","10 ms per update including image processing on a Core i7 at 2.66 GHz with single-threaded C++ in the 20 Hz real-world run (Sec. 9); estimator-only 0.93 ms per update in simulation versus 1.54 to 4.45 ms for EKF-SLAM variants (Sec. 3.4)",null,"not_applicable (no code release is stated in the paper)",[51,55],{"relation":52,"title":53,"doi_or_url":54},"conference_version","Improving the accuracy of EKF-based visual-inertial odometry (ICRA 2012, Li and Mourikis). The IJRR paper says its observability result first appeared in an earlier conference version, Li and Mourikis (2012a); the reference-list entry (ICRA, St Paul, pp. 828-835) matches this Crossref record","10.1109\u002FICRA.2012.6225229",{"relation":56,"title":57,"doi_or_url":58},"predecessor_method","A Multi-State Constraint Kalman Filter for Vision-aided Inertial Navigation (ICRA 2007)","10.1109\u002FROBOT.2007.364024",{"id":5,"kind":60,"shortName":7,"title":8,"authors":61,"year":9,"venue":64,"venueType":65,"publisher":66,"volumeIssuePages":67,"doi":68,"arxivId":48,"url":69,"firstPublicDate":70,"publicationStatus":16,"metadataStatus":71,"fulltextStatus":15,"era":10,"classicReason":72,"codeUrl":48,"cluster":11,"topics":73,"mdpi":74,"verification":75,"label":6,"fulltextRoute":76,"versionRead":77,"addedByCensus":78},"method",[62,63],"Mingyang Li","Anastasios I. Mourikis","The International Journal of Robotics Research","journal","SAGE","32(6):690-711","10.1177\u002F0278364913481251","https:\u002F\u002Fjournals.sagepub.com\u002Fdoi\u002Ffull\u002F10.1177\u002F0278364913481251","2013-06-07","metadata_verified","principle reused: shows why EKF-based VIO becomes inconsistent (yaw spuriously observable in the linearised model) and fixes it by evaluating Jacobians at first estimates plus online camera-IMU extrinsic estimation; the first-estimates idea is the FEJ treatment offered by later open filter VIO such as OpenVINS.",[11],false,"corrected","NTU institutional (Chrome)","version of record (SAGE HTML full text, IJRR 32(6), first published online 2013-06-07)",true,[80,87,91,96,100,106,110],{"category":81,"model":82,"canonical":83,"role":84,"dataset":48,"specs":85,"locator":86},"imu","Xsens MTi-G","Xsens MTI-G","method input","100 Hz","Sec. 9",{"category":88,"model":89,"canonical":89,"role":84,"dataset":48,"specs":90,"locator":86},"stereo_camera","PointGrey Bumblebee2","stereo pair, only one camera's images used, stored at 20 Hz",{"category":92,"model":93,"canonical":93,"role":94,"dataset":48,"specs":95,"locator":86},"gnss","GPS-INS (model not reported)","reference or ground truth","position ground truth",{"category":97,"model":98,"canonical":98,"role":84,"dataset":48,"specs":99,"locator":86},"platform","car (roof-mounted IMU and camera)","about 21.5 km in 37 min",{"category":81,"model":101,"canonical":101,"role":102,"dataset":103,"specs":104,"locator":105},"ISIS IMU","dataset sensor","5.5 km urban vehicle dataset used to generate simulations","not_reported","Sec. 3.4",{"category":81,"model":107,"canonical":107,"role":102,"dataset":108,"specs":85,"locator":109},"Xsens IMU (model not reported)","Cheddar Gorge dataset","Sec. 8.2",{"category":111,"model":112,"canonical":112,"role":113,"dataset":48,"specs":114,"locator":115},"compute","Intel Core i7 2.66 GHz","compute for runtime","single-threaded C++ implementation","Secs. 3.4, 9",[],1790510663209]