[{"data":1,"prerenderedAt":418},["ShallowReactive",2],{"method-blum2021precisebim":3},{"method":4,"reference":55,"equipment":76,"figures":107,"results":108},{"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":26,"sensors":34,"platform":38,"estimator":40,"association":41,"timeModel":42,"deskew":43,"loopClosure":44,"globalOptimization":44,"mapRepresentation":45,"prior":46,"outputGeometry":47,"compute":43,"codeUrl":48,"codeLicense":49,"relatedVersions":50},"blum2021precisebim","Blum et al., 2021","Localization in architectural 3D plans","Precise Robot Localization in Architectural 3D Plans",2021,"recent","C11b","localization_in_prior_map_or_bim","作者主張施工中牆體缺漏、臨時物與實作偏差使 ICP 對整棟 BIM 的對位不可靠，因此提出「局部參考」：先對整個平面圖模型做點到平面 ICP，再只對選定的參考牆面（至少三個互不平行的面）精修，並以影像密度估計網路的分數剔除或加權雜物、人員等離群點後融合到光達點。實驗在真實建築工地以靜止機器人搭配移動工人與雜物進行，並以全測站追蹤機器人上的稜鏡作為參考，且修正參考牆的竣工偏差；表 II 至 IV 的模型偏差是作者把網格上下兩側結構人為拉開 0.3 m 所模擬。","Localizes a construction robot against locally selected reference walls of an architectural plan with image-based outlier rejection, validated by total-station prism tracking on a real building site.","full_text_reviewed","peer_reviewed_published","main_body","於真實建築工地以靜止機器人測試三個位置（每處約 1 分鐘、約 300 次光達掃描，結果為三次執行的平均），現場有移動工人、木板與設備箱等雜物，並以全測站追蹤機器人上的稜鏡作為參考，且依實測參考牆偏差修正參考值。表 II 至 IV 的模型偏差是作者在網格中把上下兩側結構人為拉開 0.3 m 所模擬，並非實測施工偏差；表 I 才是未加人為偏差的雜物影響比較。屬少數在施工中環境以全測站為獨立參考、在建築平面圖模型中定位的研究（Sec. IV）。",[20,21],"real_construction_site","independent_reference",[23,24,25],"Lowest position RMSE per location came from selective localization with semantic information: 232 mm (filtered) vs 390 mm for full-model ICP at location A, 68 mm vs 222 mm at B, 52 mm (weighted) vs 76 mm at C, i.e. at least 30% lower error (Tables II-IV; Abstract)","Selective localization against reference walls constrains two directions well (trace close to the maximum eigenvalue), while full-model ICP is uncertain in more than one direction (Sec. V, Fig. 6)","Only on-board sensing, no markers or site preparation (Sec. I)",[27,28,29,30,31,32,33],"no single method combination always worked; semantic filtering performance location-dependent (Sec. V)","lateral uncertainty high where few walls constrain one axis (Sec. V)","robot stationary during evaluation; ~1 min \u002F 300 scans per location (Sec. IV-C)","Tables II-IV use a model whose upper and lower structures were moved 0.3 m apart to simulate a severe deviation; only Table I (157 to 218 mm at A and 74 to 83 mm at B with clutter) uses the unmodified plan (Sec. IV-C)","High failure rates for some variants: 76.5% (selective, weighted) and 50.3% (selective, full) at A, 52.6% (selective, weighted) at B, 25.6% (selective, filtered) at C (Tables II-IV)","Binary semantic filtering at location C removed nearly all points on the lateral reference wall; the density network, trained on NYU indoor data, partly filtered building structure outside its training domain (Sec. V)","Plan-derived mesh with a planar floor and equal wall heights because no floor or ceiling information was available (Sec. IV-B)",[35,36,37],"3D LiDAR","3 cameras","IMU",[39],"wheeled UGV (supermegabot)","per-scan registration in three steps: ICP of the scan to the full model S initialised from the previous pose, refinement by ICP to a subset R of reference surfaces (at least three mutually non-parallel surfaces; for the approximately rectangular test structures R is often the set of surfaces forming a room corner, and every test location also uses the floor as a reference surface), and rejection of the refined pose when it departs too far from the full-model result; a good initial pose (manual or from global localization) is assumed (Sec. III, III-B, Fig. 5)","point-to-plane ICP of the LiDAR scan against the building model, whose mesh is converted to a sparse point cloud; LiDAR points projected into the rectified camera images receive the per-pixel density score of the density-estimation network of Marchal et al. (trained on the NYU indoor dataset), used as a binary filter (Eq. 1) or a linear weight (Eq. 2); points outside all camera views are rejected (Sec. I, III-A)","per-scan discrete registration without a motion model; cameras and IMU hardware-synchronized with the VersaVIS trigger board, host-to-LiDAR time offset assumed negligible; evaluated only with a stationary robot (Sec. IV-A, IV-C)","not_reported","none","3D mesh generated from 2D floor plan (walls same height, planar floor) (Sec. IV-B)","architectural plan-derived mesh with selected reference surfaces","robot pose relative to plan (no map output)",null,"not_verified",[51],{"relation":52,"title":53,"doi_or_url":54},"preprint","Precise Robot Localization in Architectural 3D Plans (arXiv 2006.05137)","https:\u002F\u002Farxiv.org\u002Fabs\u002F2006.05137",{"id":5,"kind":56,"shortName":7,"title":8,"authors":57,"year":9,"venue":63,"venueType":64,"publisher":65,"volumeIssuePages":43,"doi":66,"arxivId":67,"url":54,"firstPublicDate":68,"publicationStatus":16,"metadataStatus":69,"fulltextStatus":15,"era":10,"classicReason":70,"codeUrl":48,"cluster":11,"topics":71,"mdpi":72,"verification":73,"label":6,"fulltextRoute":74,"versionRead":75,"addedByCensus":72},"method",[58,59,60,61,62],"Hermann Blum","Julian Stiefel","Cesar Cadena","Roland Siegwart","Abel Gawel","Proceedings of the 38th International Symposium on Automation and Robotics in Construction (ISARC 2021)","conference","IAARC","10.22260\u002Fisarc2021\u002F0102","2006.05137","2020-06-09","metadata_verified","not_applicable",[11],false,"confirmed","arXiv","arXiv 2006.05137v1 (9 Jun 2020, only version; carries an IEEE copyright notice for an earlier submission). The ISARC 2021 version of record (pp. 755-762) could not be read: the IAARC landing page states 'Fulltext not available (yet)'; its abstract differs only in wording (camera-based detector; error reduction of at least 30% under clutter and model deviations).",[77,83,87,92,96,101],{"category":78,"model":79,"canonical":79,"role":80,"dataset":48,"specs":81,"locator":82},"platform","supermegabot","method input","mobile robot carrying one LiDAR, three cameras and an IMU; wheeled base visible in the Fig. 4 photo; repository github.com\u002Fethz-asl\u002Feth-supermegabot (footnote 1)","Sec. IV-A, footnote 1, Fig. 4",{"category":84,"model":85,"canonical":85,"role":80,"dataset":48,"specs":43,"locator":86},"lidar","not_reported (one LiDAR, model not named)","Sec. IV-A, Fig. 4",{"category":88,"model":89,"canonical":89,"role":80,"dataset":48,"specs":90,"locator":91},"camera","not_reported (three cameras, models not named)","high field-of-view lenses; calibrated with Kalibr; one wall-facing camera per location","Sec. III-A, IV-A, Fig. 4-5",{"category":93,"model":94,"canonical":94,"role":80,"dataset":48,"specs":95,"locator":86},"imu","not_reported (IMU, model not named)","used for smooth state estimation and camera-IMU calibration",{"category":97,"model":98,"canonical":98,"role":80,"dataset":48,"specs":99,"locator":100},"other","VersaVIS camera trigger board","time-synchronizes host, cameras and IMU; LiDAR-host offset assumed negligible","Sec. IV-A",{"category":102,"model":103,"canonical":103,"role":104,"dataset":48,"specs":105,"locator":106},"total_station","not_reported (total station; model not named)","reference or ground truth","measures the position of a prism attached to the robot (the prism is written 'leica prism' in Sec. IV-C; no maker is given for the total station); referenced to the origin of the building plan; prism-to-robot offset calibrated by aligning trajectories; ground truth corrected for measured deviations of the as-built reference walls","Sec. IV-A, IV-C",[],{"totalRows":109,"groupCount":110,"groups":111,"others":417},90,3,[112,235,328],{"slug":113,"group":114,"sourceId":5,"sourceLabel":6,"table":115,"selfRows":116,"metrics":117,"seqs":133,"entrants":138,"cells":152,"outcomes":229,"locators":230,"hardware":231,"wordings":232,"notes":233},"blum2021precisebim-table-ii","blum2021precisebim:Table II","Table II",30,[118,120,122,124,126,130],{"label":119,"unit":43,"statistic":43,"alignment":44},"Pos. Repeatability, max eigenvalue of position covariance",{"label":121,"unit":43,"statistic":43,"alignment":44},"Pos. Repeatability, trace of position covariance",{"label":123,"unit":43,"statistic":43,"alignment":44},"Rot. Repeatability, max eigenvalue of rotation covariance",{"label":125,"unit":43,"statistic":43,"alignment":44},"Rot. Repeatability, trace of rotation covariance",{"label":127,"unit":128,"statistic":129,"alignment":44},"Accuracy rmse [mm] (prism position vs total station)","mm","RMSE",{"label":131,"unit":132,"statistic":43,"alignment":44},"Failure Rate [%] (scans where ICP did not converge or the selective solution was rejected)","%",[134],{"dataset":135,"sequence":136,"environment":137},"own construction-site recordings","Location A","real building construction site; stationary robot; moving worker and clutter (wooden boards, equipment boxes); mesh with a 0.3 m artificial deviation",[139,141,144,146,148,150],{"name":140,"methodId":48,"linkable":72,"proposed":72,"self":72},"full ICP, full scan",{"name":142,"methodId":5,"linkable":143,"proposed":143,"self":143},"full ICP, filtered scan",true,{"name":145,"methodId":5,"linkable":143,"proposed":143,"self":143},"full ICP, weighted scan",{"name":147,"methodId":5,"linkable":143,"proposed":143,"self":143},"selective ICP, full scan",{"name":149,"methodId":5,"linkable":143,"proposed":143,"self":143},"selective ICP, filtered scan",{"name":151,"methodId":5,"linkable":143,"proposed":143,"self":143},"selective ICP, weighted scan",[153,157,160,163,165,168,170,172,174,176,178,180,182,184,186,187,189,191,193,195,197,199,201,203,205,207,209,211,213,215,217,219,221,223,225,227],[154,154,154,155,156,154,156,156,154],0,22.3,-1,[154,158,154,159,156,154,156,156,154],1,25.9,[154,161,154,162,156,154,156,156,154],2,2.7,[154,110,154,164,156,154,156,156,154],8,[154,166,154,167,156,154,156,156,154],4,390,[154,169,154,154,156,154,156,156,154],5,[158,154,154,171,156,154,156,156,154],1.5,[158,158,154,173,156,154,156,156,154],2.5,[158,161,154,175,156,154,156,156,154],2.9,[158,110,154,177,156,154,156,156,154],8.6,[158,166,154,179,156,154,156,156,154],290,[158,169,154,181,156,154,156,156,154],4.9,[161,154,154,183,156,154,156,156,154],10.1,[161,158,154,185,156,154,156,156,154],12.7,[161,161,154,110,156,154,156,156,154],[161,110,154,188,156,154,156,156,154],8.9,[161,166,154,190,156,154,156,156,154],354,[161,169,154,192,156,154,156,156,154],8.2,[110,154,154,194,156,154,156,156,154],121.1,[110,158,154,196,156,154,156,156,154],127.2,[110,161,154,198,156,154,156,156,154],5.3,[110,110,154,200,156,154,156,156,154],15.9,[110,166,154,202,156,154,156,156,154],452,[110,169,154,204,156,154,156,156,154],50.3,[166,154,154,206,156,154,156,156,154],31.9,[166,158,154,208,156,154,156,156,154],32.2,[166,161,154,210,156,154,156,156,154],5.1,[166,110,154,212,156,154,156,156,154],15.2,[166,166,154,214,156,154,156,156,154],232,[166,169,154,216,156,154,156,156,154],44.1,[169,154,154,218,156,154,156,156,154],330.5,[169,158,154,220,156,154,156,156,154],338.6,[169,161,154,222,156,154,156,156,154],14.1,[169,110,154,224,156,154,156,156,154],42.4,[169,166,154,226,156,154,156,156,154],627,[169,169,154,228,156,154,156,156,154],76.5,[],[115],[],[],[234],"Stationary localization study: ICP against the full model or selectively against reference surfaces, using the full, semantically filtered (Eq. 1) or weighted (Eq. 2) scan; mesh with upper and lower structure moved 0.3 m apart; values averaged over three executions; repeatability units not stated",{"slug":236,"group":237,"sourceId":5,"sourceLabel":6,"table":238,"selfRows":116,"metrics":239,"seqs":246,"entrants":249,"cells":256,"outcomes":323,"locators":324,"hardware":325,"wordings":326,"notes":327},"blum2021precisebim-table-iii","blum2021precisebim:Table III","Table III",[240,241,242,243,244,245],{"label":119,"unit":43,"statistic":43,"alignment":44},{"label":121,"unit":43,"statistic":43,"alignment":44},{"label":123,"unit":43,"statistic":43,"alignment":44},{"label":125,"unit":43,"statistic":43,"alignment":44},{"label":127,"unit":128,"statistic":129,"alignment":44},{"label":131,"unit":132,"statistic":43,"alignment":44},[247],{"dataset":135,"sequence":248,"environment":137},"Location B",[250,251,252,253,254,255],{"name":140,"methodId":48,"linkable":72,"proposed":72,"self":72},{"name":142,"methodId":5,"linkable":143,"proposed":143,"self":143},{"name":145,"methodId":5,"linkable":143,"proposed":143,"self":143},{"name":147,"methodId":5,"linkable":143,"proposed":143,"self":143},{"name":149,"methodId":5,"linkable":143,"proposed":143,"self":143},{"name":151,"methodId":5,"linkable":143,"proposed":143,"self":143},[257,259,261,263,265,267,268,270,272,274,276,278,279,281,283,285,287,289,291,293,294,295,296,298,300,302,303,305,307,309,311,313,315,317,319,321],[154,154,154,258,156,154,156,156,154],3.3,[154,158,154,260,156,154,156,156,154],5.9,[154,161,154,262,156,154,156,156,154],3.2,[154,110,154,264,156,154,156,156,154],9.6,[154,166,154,266,156,154,156,156,154],222,[154,169,154,154,156,154,156,156,154],[158,154,154,269,156,154,156,156,154],1.4,[158,158,154,271,156,154,156,156,154],2.3,[158,161,154,273,156,154,156,156,154],3.4,[158,110,154,275,156,154,156,156,154],10.3,[158,166,154,277,156,154,156,156,154],342,[158,169,154,198,156,154,156,156,154],[161,154,154,280,156,154,156,156,154],3.8,[161,158,154,282,156,154,156,156,154],5.6,[161,161,154,284,156,154,156,156,154],3.6,[161,110,154,286,156,154,156,156,154],10.7,[161,166,154,288,156,154,156,156,154],230,[161,169,154,290,156,154,156,156,154],8.1,[110,154,154,292,156,154,156,156,154],7.6,[110,158,154,164,156,154,156,156,154],[110,161,154,273,156,154,156,156,154],[110,110,154,275,156,154,156,156,154],[110,166,154,297,156,154,156,156,154],328,[110,169,154,299,156,154,156,156,154],4.8,[166,154,154,301,156,154,156,156,154],2.6,[166,158,154,162,156,154,156,156,154],[166,161,154,304,156,154,156,156,154],4.2,[166,110,154,306,156,154,156,156,154],12.6,[166,166,154,308,156,154,156,156,154],68,[166,169,154,310,156,154,156,156,154],21.3,[169,154,154,312,156,154,156,156,154],17.1,[169,158,154,314,156,154,156,156,154],18.4,[169,161,154,316,156,154,156,156,154],7,[169,110,154,318,156,154,156,156,154],21,[169,166,154,320,156,154,156,156,154],244,[169,169,154,322,156,154,156,156,154],52.6,[],[238],[],[],[234],{"slug":329,"group":330,"sourceId":5,"sourceLabel":6,"table":331,"selfRows":116,"metrics":332,"seqs":339,"entrants":342,"cells":349,"outcomes":412,"locators":413,"hardware":414,"wordings":415,"notes":416},"blum2021precisebim-table-iv","blum2021precisebim:Table IV","Table IV",[333,334,335,336,337,338],{"label":119,"unit":43,"statistic":43,"alignment":44},{"label":121,"unit":43,"statistic":43,"alignment":44},{"label":123,"unit":43,"statistic":43,"alignment":44},{"label":125,"unit":43,"statistic":43,"alignment":44},{"label":127,"unit":128,"statistic":129,"alignment":44},{"label":131,"unit":132,"statistic":43,"alignment":44},[340],{"dataset":135,"sequence":341,"environment":137},"Location C",[343,344,345,346,347,348],{"name":140,"methodId":48,"linkable":72,"proposed":72,"self":72},{"name":142,"methodId":5,"linkable":143,"proposed":143,"self":143},{"name":145,"methodId":5,"linkable":143,"proposed":143,"self":143},{"name":147,"methodId":5,"linkable":143,"proposed":143,"self":143},{"name":149,"methodId":5,"linkable":143,"proposed":143,"self":143},{"name":151,"methodId":5,"linkable":143,"proposed":143,"self":143},[350,352,353,355,357,359,360,362,363,365,367,369,371,372,374,376,378,380,382,384,385,386,388,390,392,394,396,397,399,401,403,404,405,406,408,410],[154,154,154,351,156,154,156,156,154],1.8,[154,158,154,175,156,154,156,156,154],[154,161,154,354,156,154,156,156,154],4.3,[154,110,154,356,156,154,156,156,154],12.9,[154,166,154,358,156,154,156,156,154],76,[154,169,154,154,156,154,156,156,154],[158,154,154,361,156,154,156,156,154],14.8,[158,158,154,312,156,154,156,156,154],[158,161,154,364,156,154,156,156,154],4.4,[158,110,154,366,156,154,156,156,154],13.3,[158,166,154,368,156,154,156,156,154],153,[158,169,154,370,156,154,156,156,154],3.7,[161,154,154,166,156,154,156,156,154],[161,158,154,373,156,154,156,156,154],5.2,[161,161,154,375,156,154,156,156,154],4.6,[161,110,154,377,156,154,156,156,154],13.7,[161,166,154,379,156,154,156,156,154],103,[161,169,154,381,156,154,156,156,154],6.4,[110,154,154,383,156,154,156,156,154],0.9,[110,158,154,158,156,154,156,156,154],[110,161,154,354,156,154,156,156,154],[110,110,154,387,156,154,156,156,154],13,[110,166,154,389,156,154,156,156,154],65,[110,169,154,391,156,154,156,156,154],0.3,[166,154,154,393,156,154,156,156,154],28.7,[166,158,154,395,156,154,156,156,154],28.9,[166,161,154,260,156,154,156,156,154],[166,110,154,398,156,154,156,156,154],17.6,[166,166,154,400,156,154,156,156,154],154,[166,169,154,402,156,154,156,156,154],25.6,[169,154,154,158,156,154,156,156,154],[169,158,154,171,156,154,156,156,154],[169,161,154,299,156,154,156,156,154],[169,110,154,407,156,154,156,156,154],14.4,[169,166,154,409,156,154,156,156,154],52,[169,169,154,411,156,154,156,156,154],9.8,[],[331],[],[],[234],[],1790510656563]