[{"data":1,"prerenderedAt":594},["ShallowReactive",2],{"method-sgraphsplus2023":3},{"method":4,"reference":65,"equipment":89,"figures":112,"results":153},{"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":30,"sensors":35,"platform":38,"estimator":41,"association":42,"timeModel":43,"deskew":44,"loopClosure":45,"globalOptimization":46,"mapRepresentation":47,"prior":48,"outputGeometry":49,"compute":50,"codeUrl":51,"codeLicense":52,"relatedVersions":53},"sgraphsplus2023","Bavle et al., 2023","S-Graphs+","S-Graphs+: Real-Time Localization and Mapping Leveraging Hierarchical Representations",2023,"recent","C11b","full_slam_with_global_correction","S-Graphs+ 把關鍵影格位姿圖與三維場景圖放進同一個即時最佳化的因子圖，分成關鍵影格、牆面、房間與樓層四層。前端在每個新關鍵影格以序列 RANSAC 擷取牆面平面，並用以 ESDF 建立的自由空間圖分群，再與牆面組合偵測四面牆與兩面牆房間；樓層中心則由目前所有牆面中距離最寬的一組估計。後端以新的房間對牆面單一代價因子，把房間中心與所屬牆面參數一起最佳化，並沿用 S-Graphs 的掃描匹配迴圈閉合。作者在模擬環境、多處施工中住宅工地的實測資料與公開 TIERS 資料集上，以 VLP-16 資料比較軌跡誤差，以及相對於建築圖所產生三維地圖的點雲 RMSE。","LiDAR graph SLAM that jointly optimizes keyframes with a four-layer situational graph (walls, rooms, floors), using free-space-cluster room segmentation and a single room-to-wall factor; evaluated on simulated and real construction-site data (map RMSE against architectural-plan maps) and on the public TIERS dataset.","full_text_reviewed","peer_reviewed_published","main_body","作者在多處施工中住宅工地（單棟住宅兩層、四棟相連住宅三層、兩棟相連住宅兩層與一處地下室儲藏區）以機器人蒐集 VLP-16 資料，並以建築圖產生的三維地圖作為點雲 RMSE 參考（Sec. VI-A、Table II）。此參考代表設計狀態而非竣工量測，所以數值同時包含 SLAM 誤差與現場偏離圖面的部分（推論）。",[20,21,22,23],"simulation","real_construction_site","public_benchmark","cross_site",[25,26,27,28,29],"Average accuracy improvement of 10.67% over the second-best method across simulated and real experiments (Abstract; Sec. VII)","Lowest average point-cloud RMSE on the real construction-site sequences (20.9 x 10^-2 m), 5.93% better than the second-best baseline (Table II; Sec. VI-B)","Room-detection recall substantially higher than S-Graphs, adding constraints that improve accuracy in complex layouts such as C2F1 (Fig. 7; Sec. VI-B)","All modules kept real-time performance on sequences up to about 17 min (Table III; Sec. VI-B)","Same empirically chosen thresholds were used in all experiments without fine-tuning (Sec. VI-A)",[31,32,33,34],"The floor layer mostly adds semantics without significantly improving accuracy and was not ablated (Sec. VI-A)","Back-end computation time grows with sequence length as the graph grows (Sec. VI-B)","Validation on buildings with several floors and faster hierarchical optimization are left to future work (Sec. VII)","The map reference is a 3D map generated from architectural plans, not an as-built survey; C4F0 had no plan and was assessed only qualitatively (Sec. VI-A)",[36,37],"3D LiDAR (VLP-16 data used in all datasets, Sec. VI-A)","Odometry input: robot encoders on the in-house real data (the in-house platform is a legged robot, Fig. 1); LiDAR odometry (VGICP or FLOAM) on simulated and TIERS data",[39,20,40],"legged robot on the in-house construction-site data (Fig. 1; model not reported)","TIERS dataset moving platform (platform type not described in the paper)","Four-layer factor graph (keyframes, wall planes, rooms, floors) jointly optimized in real time; keyframes linked by pairwise odometry, walls by pose-plane constraints, rooms by a single room-to-wall cost per room (four-wall and two-wall variants), floors by floor-to-room relative-distance factors, plus a drift node between odometry and map frames (Sec. III, V)","Wall planes extracted from each new keyframe cloud by sequential RANSAC, converted to closest-point form in the map frame and matched to mapped planes by Mahalanobis distance (threshold 0.35 m); rooms matched by L2 distance of centres with wall-id checks (threshold 1 m); loop closures by scan matching as in S-Graphs (Sec. IV, VI-A)","discrete keyframes selected at distance-time intervals","not_reported","scan-matching loop closure module inherited from S-Graphs (Sec. III)","joint real-time optimization of keyframes, walls, rooms and floors in one factor graph (Sec. V)","keyframe point clouds with a 3D scene graph of wall planes, four-wall and two-wall rooms and floor centres; local ESDF (clearing radius 10 m) and sparse free-space graph used only for room segmentation (Sec. IV-B, VI-A)","none for estimation; architectural plans used only to build the evaluation reference","3D point-cloud map plus the four-layer situational graph; map evaluated by point-cloud RMSE against a 3D map generated from architectural plans (Table II)","Per-module mean times on construction-site sequences: plane segmentation 44.8 to 91.8 ms, back-end 74.0 to 263.1 ms, real time kept for sequences up to about 17 min (Table III); hardware not reported","https:\u002F\u002Fgithub.com\u002Fsnt-arg\u002Flidar_situational_graphs","GPL-3.0 (LICENSE file on master branch checked 2026-09-25; paper states release as a docker file)",[54,58,62],{"relation":55,"title":56,"doi_or_url":57},"preprint","S-Graphs+: Real-time Localization and Mapping leveraging Hierarchical Representations (arXiv v1 to v3)","https:\u002F\u002Farxiv.org\u002Fabs\u002F2212.11770",{"relation":59,"title":60,"doi_or_url":61},"predecessor","Situational Graphs for Robot Navigation in Structured Indoor Environments (S-Graphs), RA-L 7(4):9107-9114, 2022","https:\u002F\u002Fdoi.org\u002F10.1109\u002FLRA.2022.3189785",{"relation":63,"title":64,"doi_or_url":51},"code_release","snt-arg\u002Flidar_situational_graphs (shared S-Graphs lineage repository)",{"id":5,"kind":66,"shortName":7,"title":8,"authors":67,"year":9,"venue":73,"venueType":74,"publisher":75,"volumeIssuePages":76,"doi":77,"arxivId":78,"url":79,"firstPublicDate":80,"publicationStatus":16,"metadataStatus":81,"fulltextStatus":15,"era":10,"classicReason":82,"codeUrl":51,"cluster":11,"topics":83,"mdpi":84,"verification":85,"label":6,"fulltextRoute":86,"versionRead":87,"addedByCensus":88},"method",[68,69,70,71,72],"Hriday Bavle","Jose Luis Sanchez-Lopez","Muhammad Shaheer","Javier Civera","Holger Voos","IEEE Robotics and Automation Letters","journal","IEEE","8(8):4927-4934","10.1109\u002Flra.2023.3290512","2212.11770","https:\u002F\u002Fdoi.org\u002F10.1109\u002FLRA.2023.3290512","2022-12-22","metadata_verified","not_applicable",[11],false,"corrected","arXiv","arXiv v3 (2023-05-26, author accepted version, CC BY 4.0); IEEE Xplore version of record (RA-L 8(8), CC BY 4.0) Sec. VI text compared and found identical; VoR has no appendix",true,[90,98,103,108],{"category":91,"model":92,"canonical":93,"role":94,"dataset":95,"specs":96,"locator":97},"lidar","VLP-16","Velodyne VLP-16","method input",null,"used for all datasets; ESDF resolution derived from LiDAR resolution: 0.18 m vertical, 0.03 m horizontal","Sec. VI-A",{"category":99,"model":100,"canonical":100,"role":94,"dataset":95,"specs":101,"locator":102},"wheel_or_leg_odometry","robot encoders (model not reported)","odometry source for all in-house real sequences","Sec. VI-A (In-House Dataset)",{"category":104,"model":105,"canonical":105,"role":94,"dataset":95,"specs":106,"locator":107},"platform","legged robot (model not reported)","robot shown navigating a construction site of four adjacent houses","Fig. 1 caption",{"category":91,"model":92,"canonical":93,"role":109,"dataset":110,"specs":111,"locator":97},"dataset sensor","TIERS LiDARs dataset","TIERS multi-modal LiDAR dataset recorded by a moving platform; sequences T6-T8 small room, T10-T11 larger hallway",[113,126,135,144],{"refId":5,"refLabel":6,"fig":114,"whatZh":115,"license":116,"licenseUrl":117,"sourceUrl":118,"src":119,"width":120,"height":121,"thumb":122,"thumbWidth":123,"thumbHeight":124,"modified":125},"Fig. 1","腿式機器人（黑圈處）在四棟相連住宅施工現場建立的 S-Graph+：(a) 四層可最佳化圖的三維視圖，放大處為房間分割所用的自由空間群集；(b) 同一圖的俯視圖","CC BY 4.0","http:\u002F\u002Fcreativecommons.org\u002Flicenses\u002Fby\u002F4.0\u002F","https:\u002F\u002Far5iv.labs.arxiv.org\u002Fhtml\u002F2212.11770\u002Fassets\u002Fimages\u002Furchet1_floor_2_5.png","\u002Ffigure-files\u002Fsgraphsplus2023\u002Ffig-1.webp",599,695,"\u002Ffigure-files\u002Fsgraphsplus2023\u002Ffig-1.thumb.webp",480,557,"converted to WebP",{"refId":5,"refLabel":6,"fig":127,"whatZh":128,"license":116,"licenseUrl":117,"sourceUrl":129,"src":130,"width":131,"height":132,"thumb":133,"thumbWidth":123,"thumbHeight":134,"modified":125},"Fig. 2","S-Graphs+ 系統架構：LiDAR 與里程計輸入、牆面、房間、樓層分割與迴圈閉合前端，以及四層因子圖後端","https:\u002F\u002Far5iv.labs.arxiv.org\u002Fhtml\u002F2212.11770\u002Fassets\u002Fimages\u002Fsystem_architecture_new.png","\u002Ffigure-files\u002Fsgraphsplus2023\u002Ffig-2.webp",826,408,"\u002Ffigure-files\u002Fsgraphsplus2023\u002Ffig-2.thumb.webp",237,{"refId":5,"refLabel":6,"fig":136,"whatZh":137,"license":116,"licenseUrl":117,"sourceUrl":138,"src":139,"width":140,"height":141,"thumb":142,"thumbWidth":123,"thumbHeight":143,"modified":125},"Fig. 4 (a)","施工現場地下室序列 C4F0 由 S-Graphs+ 估計的地圖俯視圖（原圖另含 HDL-SLAM、ALOAM、FLOAM 比較）","https:\u002F\u002Far5iv.labs.arxiv.org\u002Fhtml\u002F2212.11770\u002Fassets\u002Fimages\u002Furchet2_floor0_s_graphs_blue.png","\u002Ffigure-files\u002Fsgraphsplus2023\u002Ffig-4-a.webp",888,506,"\u002Ffigure-files\u002Fsgraphsplus2023\u002Ffig-4-a.thumb.webp",274,{"refId":5,"refLabel":6,"fig":145,"whatZh":146,"license":116,"licenseUrl":117,"sourceUrl":147,"src":148,"width":149,"height":150,"thumb":151,"thumbWidth":123,"thumbHeight":152,"modified":125},"Fig. 5 (a)","施工現場序列 C2F0 由 S-Graphs+ 建立的點雲地圖（原圖另含 S-Graphs 對照）","https:\u002F\u002Far5iv.labs.arxiv.org\u002Fhtml\u002F2212.11770\u002Fassets\u002Fimages\u002Furchet1_floor0_sgraphs+.png","\u002Ffigure-files\u002Fsgraphsplus2023\u002Ffig-5-a.webp",742,381,"\u002Ffigure-files\u002Fsgraphsplus2023\u002Ffig-5-a.thumb.webp",246,{"totalRows":154,"groupCount":155,"groups":156,"others":593},72,3,[157,275,458],{"slug":158,"group":159,"sourceId":5,"sourceLabel":6,"table":160,"selfRows":161,"metrics":162,"seqs":173,"entrants":190,"cells":199,"outcomes":262,"locators":263,"hardware":265,"wordings":266,"notes":267},"sgraphsplus2023-table-iii","sgraphsplus2023:Table III","Table III",28,[163,167,169,171],{"label":164,"unit":165,"statistic":166,"alignment":82},"Computation Time (mean) [ms], module: Plane Segmentation","ms","mean",{"label":168,"unit":165,"statistic":166,"alignment":82},"Computation Time (mean) [ms], module: Room Segmentation",{"label":170,"unit":165,"statistic":166,"alignment":82},"Computation Time (mean) [ms], module: Floor Segmentation",{"label":172,"unit":165,"statistic":166,"alignment":82},"Computation Time (mean) [ms], module: Back-End",[174,178,180,182,184,186,188],{"dataset":175,"sequence":176,"environment":177},"in-house construction-site dataset (VLP-16)","C1F1","real residential construction sites, indoor floors",{"dataset":175,"sequence":179,"environment":177},"C1F2",{"dataset":175,"sequence":181,"environment":177},"C2F0",{"dataset":175,"sequence":183,"environment":177},"C2F1",{"dataset":175,"sequence":185,"environment":177},"C2F2",{"dataset":175,"sequence":187,"environment":177},"C3F1",{"dataset":175,"sequence":189,"environment":177},"C3F2",[191,193,195,197],{"name":192,"methodId":5,"linkable":88,"proposed":88,"self":88},"S-Graphs+ (Plane Segmentation)",{"name":194,"methodId":5,"linkable":88,"proposed":88,"self":88},"S-Graphs+ (Room Segmentation)",{"name":196,"methodId":5,"linkable":88,"proposed":88,"self":88},"S-Graphs+ (Floor Segmentation)",{"name":198,"methodId":5,"linkable":88,"proposed":88,"self":88},"S-Graphs+ (Back-End)",[200,204,207,210,212,215,218,221,223,225,227,229,231,233,235,237,239,241,243,245,246,248,250,252,254,256,258,260],[201,201,201,202,203,201,203,203,201],0,91.8,-1,[201,201,205,206,203,201,203,203,205],1,47.4,[201,201,208,209,203,201,203,203,208],2,68.9,[201,201,155,211,203,201,203,203,155],45.3,[201,201,213,214,203,201,203,203,213],4,82.2,[201,201,216,217,203,201,203,203,216],5,57.6,[201,201,219,220,203,201,203,203,219],6,44.8,[205,205,201,222,203,201,203,203,201],17.6,[205,205,205,224,203,201,203,203,205],9.8,[205,205,208,226,203,201,203,203,208],9.6,[205,205,155,228,203,201,203,203,155],5.3,[205,205,213,230,203,201,203,203,213],10.7,[205,205,216,232,203,201,203,203,216],2.9,[205,205,219,234,203,201,203,203,219],4.2,[208,208,201,236,203,201,203,203,201],8.1,[208,208,205,238,203,201,203,203,205],3.4,[208,208,208,240,203,201,203,203,208],4.6,[208,208,155,242,203,201,203,203,155],7.2,[208,208,213,244,203,201,203,203,213],44.7,[208,208,216,213,203,201,203,203,216],[208,208,219,247,203,201,203,203,219],16.9,[155,155,201,249,203,201,203,203,201],74,[155,155,205,251,203,201,203,203,205],105.7,[155,155,208,253,203,201,203,203,208],87.3,[155,155,155,255,203,201,203,203,155],169,[155,155,213,257,203,201,203,203,213],263.1,[155,155,216,259,203,201,203,203,216],124.5,[155,155,219,261,203,201,203,203,219],173.2,[],[264],"Table III (arXiv v3)",[],[],[268,269,270,271,272,273,274],"Mean computation time per S-Graphs+ module on the in-house construction-site sequences; sequence length 487 s for C1F1","Mean computation time per S-Graphs+ module on the in-house construction-site sequences; sequence length 657 s for C1F2","Mean computation time per S-Graphs+ module on the in-house construction-site sequences; sequence length 238 s for C2F0","Mean computation time per S-Graphs+ module on the in-house construction-site sequences; sequence length 672 s for C2F1","Mean computation time per S-Graphs+ module on the in-house construction-site sequences; sequence length 1044 s for C2F2","Mean computation time per S-Graphs+ module on the in-house construction-site sequences; sequence length 558 s for C3F1","Mean computation time per S-Graphs+ module on the in-house construction-site sequences; sequence length 999 s for C3F2",{"slug":276,"group":277,"sourceId":5,"sourceLabel":6,"table":278,"selfRows":279,"metrics":280,"seqs":285,"entrants":295,"cells":318,"outcomes":450,"locators":452,"hardware":454,"wordings":455,"notes":456},"sgraphsplus2023-table-ii","sgraphsplus2023:Table II","Table II",24,[281],{"label":282,"unit":283,"statistic":284,"alignment":44},"Point Cloud RMSE [m x 10^-2] against architectural-plan 3D map","cm","RMSE",[286,287,288,289,290,291,292,293],{"dataset":175,"sequence":176,"environment":177},{"dataset":175,"sequence":179,"environment":177},{"dataset":175,"sequence":181,"environment":177},{"dataset":175,"sequence":183,"environment":177},{"dataset":175,"sequence":185,"environment":177},{"dataset":175,"sequence":187,"environment":177},{"dataset":175,"sequence":189,"environment":177},{"dataset":175,"sequence":294,"environment":177},"Avg",[296,299,302,304,307,310,312,314,316],{"name":297,"methodId":298,"linkable":84,"proposed":84,"self":84},"HDL-SLAM [11]","koide2019_hdlgraphslam",{"name":300,"methodId":301,"linkable":88,"proposed":84,"self":84},"ALOAM [6]","aloam_software",{"name":303,"methodId":95,"linkable":84,"proposed":84,"self":84},"MLOAM [10]",{"name":305,"methodId":306,"linkable":88,"proposed":84,"self":84},"FLOAM [7]","floam2021",{"name":308,"methodId":309,"linkable":88,"proposed":84,"self":84},"LeGO-LOAM [8]","legoloam2018",{"name":311,"methodId":95,"linkable":84,"proposed":84,"self":84},"S-Graphs [9]",{"name":313,"methodId":5,"linkable":88,"proposed":88,"self":88},"S-Graphs+ w. OR",{"name":315,"methodId":5,"linkable":88,"proposed":88,"self":88},"S-Graphs+ w. OF",{"name":317,"methodId":5,"linkable":88,"proposed":88,"self":88},"S-Graphs+ (ours)",[319,321,323,325,327,329,331,333,336,338,340,342,344,346,348,350,352,354,356,358,360,362,363,364,366,368,370,372,374,376,378,380,382,383,384,385,387,388,390,392,394,396,398,400,402,403,405,407,409,410,412,414,416,418,420,421,423,425,426,428,430,431,433,435,437,440,441,442,443,444,446,448],[201,201,201,320,203,201,203,203,201],33.5,[201,201,205,322,203,201,203,203,201],19.8,[201,201,208,324,203,201,203,203,201],18.5,[201,201,155,326,203,201,203,203,201],21.1,[201,201,213,328,203,201,203,203,201],19.5,[201,201,216,330,203,201,203,203,201],22.9,[201,201,219,332,203,201,203,203,201],19.4,[201,201,334,335,203,201,203,203,201],7,22.1,[205,201,201,337,203,201,203,203,201],52.6,[205,201,205,339,203,201,203,203,201],33.6,[205,201,208,341,203,201,203,203,201],34.1,[205,201,155,343,203,201,203,203,201],45.1,[205,201,213,345,203,201,203,203,201],29.9,[205,201,216,347,203,201,203,203,201],36.5,[205,201,219,349,203,201,203,203,201],43.4,[205,201,334,351,203,201,203,203,201],39.3,[208,201,201,353,203,201,203,203,201],45,[208,201,205,355,203,201,203,203,201],27.6,[208,201,208,357,203,201,203,203,201],40.6,[208,201,155,359,203,201,203,203,201],32.4,[208,201,213,361,203,201,203,203,201],23.6,[208,201,216,95,201,201,203,203,201],[208,201,219,95,201,201,203,203,201],[208,201,334,365,203,201,203,203,201],33.8,[155,201,201,367,203,201,203,203,201],68.5,[155,201,205,369,203,201,203,203,201],39.2,[155,201,208,371,203,201,203,203,201],40.2,[155,201,155,373,203,201,203,203,201],55.5,[155,201,213,375,203,201,203,203,201],39.5,[155,201,216,377,203,201,203,203,201],58.3,[155,201,219,379,203,201,203,203,201],38.8,[155,201,334,381,203,201,203,203,201],48.6,[213,201,201,95,201,201,203,203,201],[213,201,205,95,201,201,203,203,201],[213,201,208,369,203,201,203,203,201],[213,201,155,386,203,201,203,203,201],45.5,[213,201,213,95,201,201,203,203,201],[213,201,216,389,203,201,203,203,201],52.9,[213,201,219,391,203,201,203,203,201],50.3,[213,201,334,393,203,201,203,203,201],47,[216,201,201,395,203,201,203,203,201],33.1,[216,201,205,397,203,201,203,203,201],18.9,[216,201,208,399,203,201,203,203,201],18.4,[216,201,155,401,203,201,203,203,201],21.8,[216,201,213,222,203,201,203,203,201],[216,201,216,404,203,201,203,203,201],22.8,[216,201,219,406,203,201,203,203,201],22.6,[216,201,334,408,203,201,203,203,201],22.2,[219,201,201,359,203,201,203,203,201],[219,201,205,411,203,201,203,203,201],19,[219,201,208,413,203,201,203,203,201],17.7,[219,201,155,415,203,201,203,203,201],19.9,[219,201,213,417,203,201,203,203,201],18.2,[219,201,216,419,203,201,203,203,201],24.1,[219,201,219,332,203,201,203,203,201],[219,201,334,422,203,201,203,203,201],21.5,[334,201,201,424,203,201,203,203,201],32.8,[334,201,205,411,203,201,203,203,201],[334,201,208,427,203,201,203,203,201],17,[334,201,155,429,203,201,203,203,201],20.1,[334,201,213,222,203,201,203,203,201],[334,201,216,432,203,201,203,203,201],23.3,[334,201,219,434,203,201,203,203,201],19.3,[334,201,334,436,203,201,203,203,201],21.3,[438,201,201,439,203,201,203,203,201],8,32.9,[438,201,205,397,203,201,203,203,201],[438,201,208,247,203,201,203,203,201],[438,201,155,397,203,201,203,203,201],[438,201,213,222,203,201,203,203,201],[438,201,216,445,203,201,203,203,201],22.3,[438,201,219,447,203,201,203,203,201],18.7,[438,201,334,449,203,201,203,203,201],20.9,[451],"failed (reported as unsuccessful run '-')",[453],"Table II (arXiv v3)",[],[],[457],"In-house real sequences on ongoing construction sites (C1: single house, C2: four combined houses, C3: two combined houses); all methods use odometry from robot encoders; RMSE of the estimated 3D map against the 3D map generated from the architectural plan, values in m x 10^-2; '-' marks an unsuccessful run; Avg as printed in the table",{"slug":459,"group":460,"sourceId":5,"sourceLabel":6,"table":461,"selfRows":462,"metrics":463,"seqs":466,"entrants":478,"cells":499,"outcomes":586,"locators":587,"hardware":589,"wordings":590,"notes":591},"sgraphsplus2023-table-i","sgraphsplus2023:Table I","Table I",20,[464],{"label":465,"unit":283,"statistic":44,"alignment":44},"Absolute Trajectory Error (ATE) [m x 10^-2]",[467,471,472,474,476],{"dataset":468,"sequence":469,"environment":470},"in-house simulated data (VLP-16 simulated)","C1F0","simulated indoor building floors (construction-site plans for C1F0 and C1F2)",{"dataset":468,"sequence":179,"environment":470},{"dataset":468,"sequence":473,"environment":470},"SE1",{"dataset":468,"sequence":475,"environment":470},"SE2",{"dataset":468,"sequence":477,"environment":470},"SE3",[479,481,483,485,487,489,491,493,495,497],{"name":480,"methodId":298,"linkable":84,"proposed":84,"self":84},"HDL-SLAM [11] (VGICP [26] odometry)",{"name":482,"methodId":301,"linkable":88,"proposed":84,"self":84},"ALOAM [6] (ALOAM odometry)",{"name":484,"methodId":95,"linkable":84,"proposed":84,"self":84},"MLOAM [10] (MLOAM odometry)",{"name":486,"methodId":306,"linkable":88,"proposed":84,"self":84},"FLOAM [7] (FLOAM odometry)",{"name":488,"methodId":309,"linkable":88,"proposed":84,"self":84},"LeGO-LOAM [8] (LeGO-LOAM odometry)",{"name":490,"methodId":95,"linkable":84,"proposed":84,"self":84},"S-Graphs [9] (VGICP odometry)",{"name":492,"methodId":5,"linkable":88,"proposed":88,"self":88},"S-Graphs+ w. OR (VGICP odometry; old room detection, new room-to-wall factors)",{"name":494,"methodId":5,"linkable":88,"proposed":88,"self":88},"S-Graphs+ w. OF (VGICP odometry; new room detection, old factors)",{"name":496,"methodId":5,"linkable":88,"proposed":88,"self":88},"S-Graphs+ (ours) (VGICP odometry)",{"name":498,"methodId":5,"linkable":88,"proposed":88,"self":88},"S-Graphs+ (ours) (FLOAM odometry)",[500,502,504,506,508,510,512,514,516,517,519,520,522,524,525,526,528,530,532,534,535,536,537,538,539,541,543,545,547,549,551,553,555,557,559,560,562,563,564,566,568,570,572,574,576,578,581,582,584,585],[201,201,201,501,203,201,203,203,201],9.42,[201,201,205,503,203,201,203,203,201],2.12,[201,201,208,505,203,201,203,203,201],2.46,[201,201,155,507,203,201,203,203,201],10.6,[201,201,213,509,203,201,203,203,201],6.23,[205,201,201,511,203,201,203,203,201],9.9,[205,201,205,513,203,201,203,203,201],8.7,[205,201,208,515,203,201,203,203,201],15.7,[205,201,155,371,203,201,203,203,201],[205,201,213,518,203,201,203,203,201],19.7,[208,201,201,95,201,201,203,203,201],[208,201,205,521,203,201,203,203,201],50.2,[208,201,208,523,203,201,203,203,201],66.1,[208,201,155,95,201,201,203,203,201],[208,201,213,515,203,201,203,203,201],[155,201,201,527,203,201,203,203,201],11.7,[155,201,205,529,203,201,203,203,201],14.5,[155,201,208,531,203,201,203,203,201],14.6,[155,201,155,533,203,201,203,203,201],30.5,[155,201,213,355,203,201,203,203,201],[213,201,201,95,201,201,203,203,201],[213,201,205,95,201,201,203,203,201],[213,201,208,95,201,201,203,203,201],[213,201,155,95,201,201,203,203,201],[213,201,213,540,203,201,203,203,201],74.1,[216,201,201,542,203,201,203,203,201],5.09,[216,201,205,544,203,201,203,203,201],2.57,[216,201,208,546,203,201,203,203,201],2.18,[216,201,155,548,203,201,203,203,201],9.1,[216,201,213,550,203,201,203,203,201],3.86,[219,201,201,552,203,201,203,203,201],4.95,[219,201,205,554,203,201,203,203,201],2.52,[219,201,208,556,203,201,203,203,201],1.86,[219,201,155,558,203,201,203,203,201],8.06,[219,201,213,213,203,201,203,203,201],[334,201,201,561,203,201,203,203,201],5.31,[334,201,205,155,203,201,203,203,201],[334,201,208,546,203,201,203,203,201],[334,201,155,565,203,201,203,203,201],8.4,[334,201,213,567,203,201,203,203,201],3.93,[438,201,201,569,203,201,203,203,201],4.47,[438,201,205,571,203,201,203,203,201],1.75,[438,201,208,573,203,201,203,203,201],1.91,[438,201,155,575,203,201,203,203,201],9.31,[438,201,213,577,203,201,203,203,201],3.37,[579,201,201,580,203,201,203,203,201],9,5.94,[579,201,205,527,203,201,203,203,201],[579,201,208,583,203,201,203,203,201],5.72,[579,201,155,226,203,201,203,203,201],[579,201,213,415,203,201,203,203,201],[451],[588],"Table I (arXiv v3)",[],[],[592],"Simulated experiments: C1F0 and C1F2 from 3D meshes of two floors of real architectural plans, SE1 to SE3 generic simulated indoor layouts; odometry from LiDAR only; ATE against simulator ground truth, values given in m x 10^-2; '-' marks an unsuccessful run",[],1790510662117]