[{"data":1,"prerenderedAt":794},["ShallowReactive",2],{"method-forster2017preint":3},{"method":4,"reference":66,"equipment":88,"figures":115,"results":116},{"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":24,"limitations":31,"sensors":37,"platform":40,"estimator":42,"association":43,"timeModel":44,"deskew":45,"loopClosure":46,"globalOptimization":47,"mapRepresentation":48,"prior":49,"outputGeometry":50,"compute":51,"codeUrl":52,"codeLicense":53,"relatedVersions":54},"forster2017preint","Forster et al., 2017a","On-manifold IMU preintegration","On-Manifold Preintegration for Real-Time Visual-Inertial Odometry",2017,"classic","C03","odometry_with_local_mapping","本文把兩個關鍵影格之間的大量 IMU 量測預先積分成單一相對運動約束，並提出正確處理旋轉群 SO(3) 流形結構的預積分理論，推導旋轉雜訊的性質、MAP 估計式，以及殘差、雜訊傳播與偏差事後修正的解析 Jacobian。作者將預積分 IMU 因子放入因子圖，以 iSAM2 做增量平滑，並用 structureless 視覺模型避免將三維點納入最佳化。論文亦指出其理論延伸自 Lupton 與 Sukkarieh 的原始預積分概念。","Develops IMU preintegration on the SO(3) manifold with analytic Jacobians and bias correction, and embeds it as factors in an iSAM2-based visual-inertial factor graph with structureless vision factors.","full_text_reviewed","peer_reviewed_published","main_body","not_reported。實驗包含 Vicon 室內、辦公建物周邊與多樓層建物內的手持軌跡（既有建物，非營建工地）[Sec. VIII-B]；Le Gentil 等人指出許多 LiDAR-慣性估計亦依賴預積分概念 [legentil2020gpm, Sec. I]，但本文本身未在營建場域驗證，也未評估點雲幾何。",[20,21,22,23],"simulation","controlled_experiment","independent_reference","completed_building",[25,26,27,28,29,30],"Avoids committing to a linearization point during integration compared with global-frame integration approaches (Sec. IX)","Analytic Jacobians and a-posteriori bias correction (abstract; Appendix)","iSAM2 accuracy is practically the same as batch optimization at roughly constant update time in simulation (Sec. VIII-A1)","The estimator is consistent: the average NEES over 50 Monte Carlo runs stays below the chi-square upper bound (Sec. VIII-A2)","On the 430 m indoor run, average drift over 360 m is 0.3 m versus 0.7 m for OKVIS and MSCKF (Sec. VIII-B2)","On the 160 m multi-floor run, end-to-end error is 0.5 m versus 1.4 m for Google Tango (Sec. VIII-B3)",[32,33,34,35,36],"Authors note fixed-lag smoothing window length is hard to set for guaranteed performance, motivating full smoothing (Sec. I)","Integration assumes constant orientation between IMU samples; higher-order integration is suggested for slower IMUs (Sec. V)","Biases are assumed constant between keyframes (Sec. VI, Eq. 34)","Compared systems use different front-ends, and the Tango comparison is only qualitative because the sensors differ (Sec. VIII-B2; Sec. VIII-B3)","(inference) Evaluated for monocular VIO; LiDAR-inertial use is reported only in other works",[38,39],"monocular camera (left camera of a forward-looking VI-Sensor, 20 Hz)","IMU (ADIS16448 MEMS inside the VI-Sensor, 800 Hz)",[20,41],"handheld (VI-Sensor; outdoor runs with the VI-Sensor attached to a Google Tango device and carried while walking)","factor-graph MAP estimation solved by iSAM2 incremental smoothing (full smoothing), preintegrated IMU factors on SO(3) with a-posteriori bias correction, structureless vision factors","sparse visual features tracked by the SVO front-end (sparse image alignment)","discrete keyframe states; IMU measurements preintegrated between keyframes","not_applicable","none (VIO); past states are not marginalized, so loop closures could be added (Sec. VIII-B1)","incremental smoothing over all keyframes via iSAM2","3D landmarks eliminated by structureless model (no dense map)","none","keyframe trajectory, velocities and IMU biases","real-time; average iSAM2 back-end update about 10 ms (10 optimization iterations) and SVO front-end about 3 ms per frame on a laptop with Intel i7 2.4 GHz (Sec. VIII-B, Sec. IX); full smoothing at 100 Hz claimed (Sec. I)","https:\u002F\u002Fgithub.com\u002Fborglab\u002Fgtsam","BSD (GTSAM)",[55,59,63],{"relation":56,"title":57,"doi_or_url":58},"conference_version","IMU Preintegration on Manifold for Efficient Visual-Inertial Maximum-a-Posteriori Estimation (RSS 2015)","10.15607\u002FRSS.2015.XI.006",{"relation":60,"title":61,"doi_or_url":62},"preprint","arXiv:1512.02363 (v1 2015-12-08; v3 2016-10-30)","https:\u002F\u002Farxiv.org\u002Fabs\u002F1512.02363",{"relation":64,"title":65,"doi_or_url":52},"code_release","Preintegrated IMU and structureless vision factors in GTSAM 4.0",{"id":5,"kind":67,"shortName":7,"title":68,"authors":69,"year":9,"venue":74,"venueType":75,"publisher":76,"volumeIssuePages":77,"doi":78,"arxivId":79,"url":62,"firstPublicDate":80,"publicationStatus":16,"metadataStatus":81,"fulltextStatus":15,"era":10,"classicReason":82,"codeUrl":52,"cluster":11,"topics":83,"mdpi":84,"verification":85,"label":6,"fulltextRoute":86,"versionRead":87,"addedByCensus":84},"method","On-Manifold Preintegration for Real-Time Visual–Inertial Odometry",[70,71,72,73],"Christian Forster","Luca Carlone","Frank Dellaert","Davide Scaramuzza","IEEE Transactions on Robotics","journal","IEEE","33(1):1-21","10.1109\u002Ftro.2016.2597321","1512.02363","2015-07-13","metadata_verified","principle reused: on-manifold IMU preintegration factors (first public at RSS 2015 and arXiv 2015-12) are reused in factor-graph visual-inertial and LiDAR-inertial back-ends; necessary to explain IMU handling in smoothing-based systems.",[11],false,"corrected","arXiv","arXiv 1512.02363v3 (2016-10-30), post-acceptance version stating acceptance in IEEE T-RO with DOI 10.1109\u002FTRO.2016.2597321; the IEEE version of record was not opened",[89,96,100,105,110],{"category":90,"model":91,"canonical":91,"role":92,"dataset":93,"specs":94,"locator":95},"imu","ADIS16448","method input",null,"MEMS IMU in the VI-Sensor, 800 Hz","Sec. VIII-B2",{"category":97,"model":98,"canonical":98,"role":92,"dataset":93,"specs":99,"locator":95},"camera","VI-Sensor","forward-looking; two embedded WVGA monochrome cameras, only the left camera used; 20 Hz",{"category":101,"model":102,"canonical":102,"role":103,"dataset":93,"specs":104,"locator":95},"other","Vicon system","reference or ground truth","mounted in the room; hand-eye calibration to the camera by least squares",{"category":101,"model":106,"canonical":106,"role":107,"dataset":93,"specs":108,"locator":109},"Google Tango Peanut sensor (mapper version 3.15)","compared device","engineered VIO device; VI-Sensor rigidly attached to it for the outdoor runs","Sec. VIII-B3",{"category":111,"model":112,"canonical":112,"role":113,"dataset":93,"specs":114,"locator":95},"compute","standard laptop (Intel i7, 2.4 GHz)","compute for runtime","Intel i7, 2.4 GHz",[],{"totalRows":117,"groupCount":118,"groups":119,"others":741},98,13,[120,361,521,660],{"slug":121,"group":122,"sourceId":123,"sourceLabel":124,"table":125,"selfRows":126,"metrics":127,"seqs":135,"entrants":160,"cells":171,"outcomes":355,"locators":356,"hardware":357,"wordings":358,"notes":359},"eckenhoff2019closedform-table-ii","eckenhoff2019closedform:Table II","eckenhoff2019closedform","Eckenhoff et al., 2019","Table II",22,[128,132],{"label":129,"unit":130,"statistic":131,"alignment":49},"position RMSE","m","RMSE",{"label":133,"unit":134,"statistic":131,"alignment":49},"orientation RMSE","deg",[136,140,142,144,146,148,150,152,154,156,158],{"dataset":137,"sequence":138,"environment":139},"EuRoC MAV","V1 01 easy","public benchmark (EuRoC MAV)",{"dataset":137,"sequence":141,"environment":139},"V1 02 med",{"dataset":137,"sequence":143,"environment":139},"V1 03 diff",{"dataset":137,"sequence":145,"environment":139},"V2 01 easy",{"dataset":137,"sequence":147,"environment":139},"V2 02 med",{"dataset":137,"sequence":149,"environment":139},"V2 03 diff",{"dataset":137,"sequence":151,"environment":139},"MH 01 easy",{"dataset":137,"sequence":153,"environment":139},"MH 02 easy",{"dataset":137,"sequence":155,"environment":139},"MH 03 med",{"dataset":137,"sequence":157,"environment":139},"MH 04 diff",{"dataset":137,"sequence":159,"environment":139},"MH 05 diff",[161,164,166,168],{"name":162,"methodId":123,"linkable":163,"proposed":163,"self":84},"MODEL-1",true,{"name":165,"methodId":123,"linkable":163,"proposed":163,"self":84},"MODEL-2",{"name":167,"methodId":5,"linkable":163,"proposed":84,"self":163},"DISCRETE",{"name":169,"methodId":170,"linkable":163,"proposed":84,"self":84},"OKVIS","okvis2015",[172,176,179,181,183,186,188,191,193,195,197,199,201,203,205,207,209,211,213,215,217,219,221,223,225,227,229,231,233,234,236,238,240,243,245,247,249,251,253,255,257,260,262,264,266,268,270,272,274,277,279,281,283,285,287,289,291,294,296,298,300,302,304,305,306,309,311,313,315,316,318,320,322,325,327,329,331,333,335,337,339,342,344,346,348,350,351,353],[173,173,173,174,175,173,175,175,173],0,0.2522,-1,[173,177,173,178,175,173,175,175,173],1,2.749,[177,173,173,180,175,173,175,175,173],0.216,[177,177,173,182,175,173,175,175,173],2.503,[184,173,173,185,175,173,175,175,173],2,0.2547,[184,177,173,187,175,173,175,175,173],2.781,[189,173,173,190,175,173,175,175,173],3,0.2356,[189,177,173,192,175,173,175,175,173],2.458,[173,173,177,194,175,173,175,175,173],0.1342,[173,177,177,196,175,173,175,175,173],0.942,[177,173,177,198,175,173,175,175,173],0.1214,[177,177,177,200,175,173,175,175,173],1.215,[184,173,177,202,175,173,175,175,173],0.1344,[184,177,177,204,175,173,175,175,173],1.001,[189,173,177,206,175,173,175,175,173],0.1996,[189,177,177,208,175,173,175,175,173],2.321,[173,173,184,210,175,173,175,175,173],0.1101,[173,177,184,212,175,173,175,175,173],0.88,[177,173,184,214,175,173,175,175,173],0.0953,[177,177,184,216,175,173,175,175,173],0.809,[184,173,184,218,175,173,175,175,173],0.1012,[184,177,184,220,175,173,175,175,173],0.83,[189,173,184,222,175,173,175,175,173],0.183,[189,177,184,224,175,173,175,175,173],3.498,[173,173,189,226,175,173,175,175,173],0.1429,[173,177,189,228,175,173,175,175,173],1.069,[177,173,189,230,175,173,175,175,173],0.1426,[177,177,189,232,175,173,175,175,173],1.148,[184,173,189,230,175,173,175,175,173],[184,177,189,235,175,173,175,175,173],1.118,[189,173,189,237,175,173,175,175,173],0.1806,[189,177,189,239,175,173,175,175,173],0.973,[173,173,241,242,175,173,175,175,173],4,0.1297,[173,177,241,244,175,173,175,175,173],1.39,[177,173,241,246,175,173,175,175,173],0.1223,[177,177,241,248,175,173,175,175,173],1.135,[184,173,241,250,175,173,175,175,173],0.1375,[184,177,241,252,175,173,175,175,173],1.45,[189,173,241,254,175,173,175,175,173],0.1695,[189,177,241,256,175,173,175,175,173],2.334,[173,173,258,259,175,173,175,175,173],5,0.2982,[173,177,258,261,175,173,175,175,173],2.159,[177,173,258,263,175,173,175,175,173],0.28,[177,177,258,265,175,173,175,175,173],1.769,[184,173,258,267,175,173,175,175,173],0.3055,[184,177,258,269,175,173,175,175,173],2.052,[189,173,258,271,175,173,175,175,173],0.3483,[189,177,258,273,175,173,175,175,173],8.327,[173,173,275,276,175,173,175,175,173],6,0.1817,[173,177,275,278,175,173,175,175,173],1.398,[177,173,275,280,175,173,175,175,173],0.1653,[177,177,275,282,175,173,175,175,173],1.761,[184,173,275,284,175,173,175,175,173],0.205,[184,177,275,286,175,173,175,175,173],1.321,[189,173,275,288,175,173,175,175,173],0.2523,[189,177,275,290,175,173,175,175,173],0.728,[173,173,292,293,175,173,175,175,173],7,0.1533,[173,177,292,295,175,173,175,175,173],0.691,[177,173,292,297,175,173,175,175,173],0.1498,[177,177,292,299,175,173,175,175,173],0.525,[184,173,292,301,175,173,175,175,173],0.1564,[184,177,292,303,175,173,175,175,173],0.599,[189,173,292,288,175,173,175,175,173],[189,177,292,290,175,173,175,175,173],[173,173,307,308,175,173,175,175,173],8,0.2993,[173,177,307,310,175,173,175,175,173],1.024,[177,173,307,312,175,173,175,175,173],0.2627,[177,177,307,314,175,173,175,175,173],0.968,[184,173,307,263,175,173,175,175,173],[184,177,307,317,175,173,175,175,173],0.84,[189,173,307,319,175,173,175,175,173],0.3193,[189,177,307,321,175,173,175,175,173],1.903,[173,173,323,324,175,173,175,175,173],9,0.3312,[173,177,323,326,175,173,175,175,173],0.849,[177,173,323,328,175,173,175,175,173],0.3515,[177,177,323,330,175,173,175,175,173],0.974,[184,173,323,332,175,173,175,175,173],0.3488,[184,177,323,334,175,173,175,175,173],0.852,[189,173,323,336,175,173,175,175,173],0.2145,[189,177,323,338,175,173,175,175,173],1.022,[173,173,340,341,175,173,175,175,173],10,0.3939,[173,177,340,343,175,173,175,175,173],0.692,[177,173,340,345,175,173,175,175,173],0.3971,[177,177,340,347,175,173,175,175,173],0.715,[184,173,340,349,175,173,175,175,173],0.3835,[184,177,340,216,175,173,175,175,173],[189,173,340,352,175,173,175,175,173],0.5432,[189,177,340,354,175,173,175,175,173],0.738,[],[125],[],[],[360],"Indirect stereo VIO; absolute RMSE averaged over 10 runs; ground-truth initialisation",{"slug":362,"group":363,"sourceId":123,"sourceLabel":124,"table":364,"selfRows":126,"metrics":365,"seqs":368,"entrants":380,"cells":384,"outcomes":515,"locators":516,"hardware":517,"wordings":518,"notes":519},"eckenhoff2019closedform-table-iv","eckenhoff2019closedform:Table IV","Table IV",[366,367],{"label":129,"unit":130,"statistic":131,"alignment":49},{"label":133,"unit":134,"statistic":131,"alignment":49},[369,370,371,372,373,374,375,376,377,378,379],{"dataset":137,"sequence":138,"environment":139},{"dataset":137,"sequence":141,"environment":139},{"dataset":137,"sequence":143,"environment":139},{"dataset":137,"sequence":145,"environment":139},{"dataset":137,"sequence":147,"environment":139},{"dataset":137,"sequence":149,"environment":139},{"dataset":137,"sequence":151,"environment":139},{"dataset":137,"sequence":153,"environment":139},{"dataset":137,"sequence":155,"environment":139},{"dataset":137,"sequence":157,"environment":139},{"dataset":137,"sequence":159,"environment":139},[381,382,383],{"name":162,"methodId":123,"linkable":163,"proposed":163,"self":84},{"name":165,"methodId":123,"linkable":163,"proposed":163,"self":84},{"name":167,"methodId":5,"linkable":163,"proposed":84,"self":163},[385,387,389,391,393,395,397,399,401,403,405,407,409,411,413,415,417,418,420,422,424,426,428,430,432,434,436,438,440,442,444,446,448,450,452,454,456,458,460,462,464,466,468,470,472,474,476,478,479,481,483,485,487,489,491,493,495,497,499,501,503,505,507,509,511,513],[173,173,173,386,175,173,175,175,173],0.2445,[173,177,173,388,175,173,175,175,173],2.218,[177,173,173,390,175,173,175,175,173],0.2482,[177,177,173,392,175,173,175,175,173],2.246,[184,173,173,394,175,173,175,175,173],0.253,[184,177,173,396,175,173,175,175,173],2.223,[173,173,177,398,175,173,175,175,173],0.1598,[173,177,177,400,175,173,175,175,173],1.767,[177,173,177,402,175,173,175,175,173],0.1309,[177,177,177,404,175,173,175,175,173],1.483,[184,173,177,406,175,173,175,175,173],0.1763,[184,177,177,408,175,173,175,175,173],1.899,[173,173,184,410,175,173,175,175,173],0.099,[173,177,184,412,175,173,175,175,173],1.18,[177,173,184,414,175,173,175,175,173],0.103,[177,177,184,416,175,173,175,175,173],1.279,[184,173,184,414,175,173,175,175,173],[184,177,184,419,175,173,175,175,173],1.234,[173,173,189,421,175,173,175,175,173],0.1627,[173,177,189,423,175,173,175,175,173],2.089,[177,173,189,425,175,173,175,175,173],0.194,[177,177,189,427,175,173,175,175,173],1.956,[184,173,189,429,175,173,175,175,173],0.1664,[184,177,189,431,175,173,175,175,173],1.533,[173,173,241,433,175,173,175,175,173],0.1809,[173,177,241,435,175,173,175,175,173],2.53,[177,173,241,437,175,173,175,175,173],0.1665,[177,177,241,439,175,173,175,175,173],2.527,[184,173,241,441,175,173,175,175,173],0.1688,[184,177,241,443,175,173,175,175,173],2.309,[173,173,258,445,175,173,175,175,173],0.9337,[173,177,258,447,175,173,175,175,173],6.187,[177,173,258,449,175,173,175,175,173],0.8927,[177,177,258,451,175,173,175,175,173],5.425,[184,173,258,453,175,173,175,175,173],1.0137,[184,177,258,455,175,173,175,175,173],4.998,[173,173,275,457,175,173,175,175,173],0.2947,[173,177,275,459,175,173,175,175,173],2.27,[177,173,275,461,175,173,175,175,173],0.3277,[177,177,275,463,175,173,175,175,173],2.148,[184,173,275,465,175,173,175,175,173],0.3217,[184,177,275,467,175,173,175,175,173],2.226,[173,173,292,469,175,173,175,175,173],0.1882,[173,177,292,471,175,173,175,175,173],1.65,[177,173,292,473,175,173,175,175,173],0.2136,[177,177,292,475,175,173,175,175,173],1.582,[184,173,292,477,175,173,175,175,173],0.2008,[184,177,292,404,175,173,175,175,173],[173,173,307,480,175,173,175,175,173],0.233,[173,177,307,482,175,173,175,175,173],2.096,[177,173,307,484,175,173,175,175,173],0.2295,[177,177,307,486,175,173,175,175,173],2.121,[184,173,307,488,175,173,175,175,173],0.2288,[184,177,307,490,175,173,175,175,173],2.092,[173,173,323,492,175,173,175,175,173],0.4792,[173,177,323,494,175,173,175,175,173],2.513,[177,173,323,496,175,173,175,175,173],0.4867,[177,177,323,498,175,173,175,175,173],2.627,[184,173,323,500,175,173,175,175,173],0.4724,[184,177,323,502,175,173,175,175,173],2.562,[173,173,340,504,175,173,175,175,173],0.2884,[173,177,340,506,175,173,175,175,173],1.664,[177,173,340,508,175,173,175,175,173],0.3014,[177,177,340,510,175,173,175,175,173],1.722,[184,173,340,512,175,173,175,175,173],0.2946,[184,177,340,514,175,173,175,175,173],1.7,[],[364],[],[],[520],"Direct stereo VINS (iSAM2, loop closures); absolute RMSE averaged over 10 runs; ground-truth initialisation",{"slug":522,"group":523,"sourceId":524,"sourceLabel":525,"table":125,"selfRows":526,"metrics":527,"seqs":537,"entrants":555,"cells":563,"outcomes":654,"locators":655,"hardware":656,"wordings":657,"notes":658},"legentil2020gpm-table-ii","legentil2020gpm:Table II","legentil2020gpm","Le Gentil et al., 2020",16,[528,532,535],{"label":529,"unit":530,"statistic":531,"alignment":49},"average absolute rotation error of preintegrated measurement","mrad","mean",{"label":533,"unit":534,"statistic":531,"alignment":49},"average absolute position error of preintegrated measurement","mm",{"label":529,"unit":536,"statistic":531,"alignment":49},"rad (as printed; the 1D block uses mrad)",[538,541,543,545,547,549,551,553],{"dataset":539,"sequence":540,"environment":20},"simulated IMU trajectories","1D rotations, query rate 20 Hz",{"dataset":539,"sequence":542,"environment":20},"1D rotations, query rate 10 Hz",{"dataset":539,"sequence":544,"environment":20},"1D rotations, query rate 2 Hz",{"dataset":539,"sequence":546,"environment":20},"1D rotations, query rate 1 Hz",{"dataset":539,"sequence":548,"environment":20},"3D rotations, query rate 20 Hz",{"dataset":539,"sequence":550,"environment":20},"3D rotations, query rate 10 Hz",{"dataset":539,"sequence":552,"environment":20},"3D rotations, query rate 2 Hz",{"dataset":539,"sequence":554,"environment":20},"3D rotations, query rate 1 Hz",[556,558,561],{"name":557,"methodId":5,"linkable":163,"proposed":84,"self":163},"PM (on-manifold preintegration [14])",{"name":559,"methodId":560,"linkable":163,"proposed":84,"self":84},"UPM (upsampled preintegration [17])","legentil2018lidarimucalib",{"name":562,"methodId":524,"linkable":163,"proposed":163,"self":84},"GPM",[564,566,568,569,571,573,575,577,579,581,583,585,587,589,591,593,595,597,599,601,603,605,607,608,610,612,614,616,618,619,620,622,624,626,627,629,631,633,635,637,639,641,643,645,647,649,651,652],[173,173,173,565,175,173,175,175,173],2.41,[173,177,173,567,175,173,175,175,173],5.61,[177,173,173,263,175,173,175,175,173],[177,177,173,570,175,173,175,175,173],0.6,[184,173,173,572,175,173,175,175,173],0.1,[184,177,173,574,175,173,175,175,173],0.02,[173,173,177,576,175,173,175,175,173],4.91,[173,177,177,578,175,173,175,175,173],12.1,[177,173,177,580,175,173,175,175,173],1.61,[177,177,177,582,175,173,175,175,173],1.32,[184,173,177,584,175,173,175,175,173],1.4,[184,177,177,586,175,173,175,175,173],0.16,[173,173,184,588,175,173,175,175,173],21.3,[173,177,184,590,175,173,175,175,173],61.3,[177,173,184,592,175,173,175,175,173],3.09,[177,177,184,594,175,173,175,175,173],7.19,[184,173,184,596,175,173,175,175,173],0.2,[184,177,184,598,175,173,175,175,173],0.61,[173,173,189,600,175,173,175,175,173],24.9,[173,177,189,602,175,173,175,175,173],125,[177,173,189,604,175,173,175,175,173],2.51,[177,177,189,606,175,173,175,175,173],12.6,[184,173,189,263,175,173,175,175,173],[184,177,189,609,175,173,175,175,173],1.8,[173,184,241,611,175,173,175,175,173],5.48,[173,177,241,613,175,173,175,175,173],6.06,[177,184,241,615,175,173,175,175,173],0.8,[177,177,241,617,175,173,175,175,173],0.65,[184,184,241,615,175,173,175,175,173],[184,177,241,574,175,173,175,175,173],[173,184,258,621,175,173,175,175,173],11.1,[173,177,258,623,175,173,175,175,173],13.1,[177,184,258,625,175,173,175,175,173],7.53,[177,177,258,459,175,173,175,175,173],[184,184,258,628,175,173,175,175,173],7.18,[184,177,258,630,175,173,175,175,173],0.48,[173,184,275,632,175,173,175,175,173],35.6,[173,177,275,634,175,173,175,175,173],65,[177,184,275,636,175,173,175,175,173],4.72,[177,177,275,638,175,173,175,175,173],7.29,[184,184,275,640,175,173,175,175,173],3.56,[184,177,275,642,175,173,175,175,173],1.74,[173,184,292,644,175,173,175,175,173],37.5,[173,177,292,646,175,173,175,175,173],159,[177,184,292,648,175,173,175,175,173],3.75,[177,177,292,650,175,173,175,175,173],15.9,[184,184,292,648,175,173,175,175,173],[184,177,292,653,175,173,175,175,173],10.3,[],[125],[],[],[659],"Simulated fast trajectories, 100 trials, fixed query rates of 1 to 20 Hz; average absolute error of the preintegrated measurement (the paper calls it absolute pose error); units as printed: mrad and mm for 1D rotations, 'rad' and mm for 3D rotations",{"slug":661,"group":662,"sourceId":524,"sourceLabel":525,"table":663,"selfRows":526,"metrics":664,"seqs":668,"entrants":701,"cells":703,"outcomes":734,"locators":735,"hardware":736,"wordings":738,"notes":739},"legentil2020gpm-table-iii","legentil2020gpm:Table III","Table III",[665],{"label":666,"unit":667,"statistic":531,"alignment":45},"average computation time per preintegrated measurement","ms",[669,671,673,675,677,679,681,683,685,687,689,691,693,695,697,699],{"dataset":539,"sequence":670,"environment":20},"1D rotations, interval 0.05 s",{"dataset":539,"sequence":672,"environment":20},"1D rotations, interval 0.1 s",{"dataset":539,"sequence":674,"environment":20},"1D rotations, interval 0.5 s",{"dataset":539,"sequence":676,"environment":20},"1D rotations, interval 1 s",{"dataset":539,"sequence":678,"environment":20},"1D rotations, interval 2 s",{"dataset":539,"sequence":680,"environment":20},"1D rotations, interval 3 s",{"dataset":539,"sequence":682,"environment":20},"1D rotations, interval 4 s",{"dataset":539,"sequence":684,"environment":20},"1D rotations, interval 5 s",{"dataset":539,"sequence":686,"environment":20},"3D rotations, interval 0.05 s",{"dataset":539,"sequence":688,"environment":20},"3D rotations, interval 0.1 s",{"dataset":539,"sequence":690,"environment":20},"3D rotations, interval 0.5 s",{"dataset":539,"sequence":692,"environment":20},"3D rotations, interval 1 s",{"dataset":539,"sequence":694,"environment":20},"3D rotations, interval 2 s",{"dataset":539,"sequence":696,"environment":20},"3D rotations, interval 3 s",{"dataset":539,"sequence":698,"environment":20},"3D rotations, interval 4 s",{"dataset":539,"sequence":700,"environment":20},"3D rotations, interval 5 s",[702],{"name":557,"methodId":5,"linkable":163,"proposed":84,"self":163},[704,705,707,709,711,713,715,717,719,720,721,722,724,727,729,732],[173,173,173,615,175,173,173,175,173],[173,173,177,706,175,173,173,175,173],0.9,[173,173,184,708,175,173,173,175,173],3.1,[173,173,189,710,175,173,173,175,173],5.8,[173,173,241,712,175,173,173,175,173],11.6,[173,173,258,714,175,173,173,175,173],18.6,[173,173,275,716,175,173,173,175,173],23.3,[173,173,292,718,175,173,173,175,173],28.5,[173,173,307,570,175,173,173,175,173],[173,173,323,706,175,173,173,175,173],[173,173,340,708,175,173,173,175,173],[173,173,723,710,175,173,173,175,173],11,[173,173,725,726,175,173,173,175,173],12,11.8,[173,173,118,728,175,173,173,175,173],17.5,[173,173,730,731,175,173,173,175,173],14,24.1,[173,173,733,718,175,173,173,175,173],15,[],[663],[737],"laptop with Intel i5-6300U CPU at 2.40 GHz and 24 GiB RAM; single-thread Matlab code, not optimised",[],[740],"Average computation time over 50 trials for different integration interval lengths; IMU 100 Hz, UPM upsampled to 1 kHz; for UPM and GPM the hyper-parameter training time and the inference time are listed separately",[742,747,753,759,765,771,777,782,788],{"group":743,"slug":744,"sourceLabel":525,"table":745,"selfRows":307,"datasets":746},"legentil2020gpm:Table I","legentil2020gpm-table-i","Table I",[539],{"group":748,"slug":749,"sourceLabel":750,"table":751,"selfRows":275,"datasets":752},"d3vo2020:Table 6","d3vo2020-table-6","Yang et al., 2020a","Table 6",[137],{"group":754,"slug":755,"sourceLabel":6,"table":756,"selfRows":184,"datasets":757},"forster2017preint:Text Sec.VIII-B2 (timing)","forster2017preint-text-sec-viii-b2-timing","Text Sec.VIII-B2 (timing)",[758],"430 m indoor sequence recorded with a forward-looking VI-Sensor, with Vicon ground truth; dataset and OKVIS and MSCKF trajectories obtained from the OKVIS authors",{"group":760,"slug":761,"sourceLabel":124,"table":762,"selfRows":177,"datasets":763},"eckenhoff2019closedform:Text Sec.VII-A2 (Gore Hall)","eckenhoff2019closedform-text-sec-vii-a2-gore-hall","Text Sec.VII-A2 (Gore Hall)",[764],"UD Gore Hall",{"group":766,"slug":767,"sourceLabel":124,"table":768,"selfRows":177,"datasets":769},"eckenhoff2019closedform:Text Sec.VII-A2 (Smith Hall)","eckenhoff2019closedform-text-sec-vii-a2-smith-hall","Text Sec.VII-A2 (Smith Hall)",[770],"UD Smith Hall",{"group":772,"slug":773,"sourceLabel":6,"table":774,"selfRows":177,"datasets":775},"forster2017preint:Text Sec.VIII-A1","forster2017preint-text-sec-viii-a1","Text Sec.VIII-A1",[776],"simulation (circular trajectory with sinusoidal vertical motion)",{"group":778,"slug":779,"sourceLabel":6,"table":780,"selfRows":177,"datasets":781},"forster2017preint:Text Sec.VIII-B2 (drift)","forster2017preint-text-sec-viii-b2-drift","Text Sec.VIII-B2 (drift)",[758],{"group":783,"slug":784,"sourceLabel":6,"table":785,"selfRows":177,"datasets":786},"forster2017preint:Text Sec.VIII-B3 (multi-floor)","forster2017preint-text-sec-viii-b3-multi-floor","Text Sec.VIII-B3 (multi-floor)",[787],"own indoor multi-floor sequence",{"group":789,"slug":790,"sourceLabel":6,"table":791,"selfRows":177,"datasets":792},"forster2017preint:Text Sec.VIII-B3 (outdoor loop)","forster2017preint-text-sec-viii-b3-outdoor-loop","Text Sec.VIII-B3 (outdoor loop)",[793],"own outdoor sequence",1790510659284]