[{"data":1,"prerenderedAt":236},["ShallowReactive",2],{"method-vilslam2019":3},{"method":4,"reference":59,"equipment":82,"figures":118,"results":119},{"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":36,"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},"vilslam2019","Shao et al., 2019","VIL-SLAM","Stereo Visual Inertial LiDAR Simultaneous Localization and Mapping",2019,"recent","C07","full_slam_with_global_correction","VIL-SLAM 把三個模組串接：緊耦合的雙目視覺慣性里程計以固定滯後的位姿圖平滑器估計運動，並以 IMU 頻率輸出位姿；LiDAR 建圖模組用這些位姿為每個點去畸變，再以 LOAM 式邊緣與平面特徵做掃描對地圖配準；迴圈閉合先以視覺詞袋偵測候選並用 EPnP 求初始約束，再以稀疏 LiDAR 特徵點的 ICP 精修，最後用 iSAM2 增量最佳化全域位姿圖，並把修正後的位姿即時回饋給建圖模組重新定位。作者以 Faro 地面雷射掃描為參考評估地圖，並在隧道與走廊等 LiDAR 退化場景顯示視覺慣性先驗的幫助。","Loosely chained stereo VIO (tightly coupled fixed-lag smoother), LOAM-style LiDAR mapping seeded and dewarped by IMU-rate VIO poses, and LiDAR-enhanced visual loop closure (BoW detection, EPnP, sparse-feature ICP) with incremental iSAM2 pose-graph optimization, targeting LiDAR-degenerate tunnels and hallways.","full_text_reviewed","peer_reviewed_published","supplementary","測試場景為倉庫式高挑空間、無特徵走廊、隧道與戶外道路，未在施工中工地；地圖精度以 Faro 飛時測距雷射掃描為參考，先對齊再計算地圖點到參考模型最近點的平均距離，屬於有獨立幾何參考的點雲評估，這種以 TLS 為真值的做法可直接借鏡到施工點雲驗收。走廊與隧道的 LiDAR 退化問題也常見於施工中的地下或長廊空間（推論）。",[20,21,22,23],"completed_building","underground_or_tunnel","independent_reference","public_benchmark",[25,26,27,28,29],"Succeeded in the hallway and tunnel tests, where LOAM accumulated large error or failed because of degeneracy along the traversal direction (Sec. VIII-B; Table I)","Lower final drift error than LOAM on highbay and hallway and equal on outdoor; lower mean map registration error than LOAM against Faro scans on highbay, hallway and huge loop (Table I)","Loop closure lowered the final drift to 0.05% in the hallway and 0.08% in the tunnel (Sec. VIII-B)","LOAM failed the huge-loop test after re-entering the highbay, while VIL-SLAM reached 0.01% final drift without a loop closure being triggered (Sec. VIII-B)","Stereo VIO succeeded on all EuRoC sequences with accuracy comparable to three state-of-the-art methods (Sec. VIII-C; Fig. 8)",[31,32,33,34,35],"In the tunnel, degeneracy still caused error along the traversal direction; a single loop constraint did not fully remove the doubled map (Sec. VIII-B; Fig. 7)","Featureless hallway walls under-constrain the VIO and misalign the map until loop closure (Sec. VIII-B)","VIO and LiDAR mapping are only loosely coupled, so LiDAR does not correct IMU biases (Sec. IX)","Loop-closure ICP uses sparse features between scans; the authors expect scan-to-map matching would give better constraints (Sec. IX)","Only compared with one LiDAR baseline (LOAM) on custom datasets; KITTI not used because the evaluation sequences lack inertial data (Sec. VIII)",[37,38,39],"stereo camera pair (two megapixel cameras; model not reported)","16 scan-line 3D LiDAR (model not reported)","IMU at 400 Hz (model not reported)",[41],"not_reported (custom sensor platform shown in Fig. 1(a); carrier platform not stated)","loosely coupled chain: tightly coupled stereo VIO as a fixed-lag smoother over the most recent N stereo frames (IMU pre-integration factors and structureless vision factors, Levenberg-Marquardt, Schur-complement marginalization, GTSAM) provides IMU-rate motion priors to LOAM-style LiDAR mapping; a global pose graph of LiDAR mapping poses with LiDAR odometry and loop constraint factors is optimized incrementally with iSAM2 (Secs. V-VII)","visual: KLT tracking of stereo matches, Shi-Tomasi corners with ORB descriptors and brute-force stereo matching; LiDAR: edge and planar feature points registered scan-to-map by point-to-line (two closest edge points) and point-to-plane (three closest surface points) distances as in LOAM (Secs. IV, VI-B)","discrete stereo-frame states in the VIO; LiDAR points dewarped with IMU-rate VIO poses; custom microcontroller circuit synchronizes cameras, LiDAR, IMU and computer by simulating GPS time signals (Secs. V, VI-A, VIII-A)","each LiDAR point dewarped to the end-of-scan time using the closest IMU-rate VIO poses (Sec. VI-A, Eq. 6)","yes; visual Bag-of-Words detection (DBoW3) of key images within a time threshold, descriptor matching to reject false positives, EPnP initial constraint, then ICP refinement on sparse LiDAR feature points of the key scans (LibPointMatcher) (Sec. VII-A, VII-B)","incremental global pose-graph optimization with iSAM2 over all LiDAR mapping poses; corrected poses sent back in real time so LiDAR mapping re-localizes and updates its feature map (Sec. VII-C, VII-D)","sparse LiDAR feature map (all previous edge and surface feature points) for registration; dense map produced in post-processing by stitching dewarped scans with the best estimated poses, reported as 1 cm voxel dense maps near real time (Sec. III; abstract)","none","loop-closure-corrected 6-DoF LiDAR poses in real time and a dense point-cloud map near real time (abstract; Fig. 5)","4 GHz computer with 4 physical cores on the platform; EuRoC VIO results obtained in real time on a desktop with a 3.60 GHz i7-4790 CPU (Secs. VIII-A, VIII-C); no per-module timing reported",null,"not_verified",[55],{"relation":56,"title":57,"doi_or_url":58},"preprint","Stereo Visual Inertial LiDAR Simultaneous Localization and Mapping (arXiv v1, submitted version)","https:\u002F\u002Farxiv.org\u002Fabs\u002F1902.10741",{"id":5,"kind":60,"shortName":7,"title":8,"authors":61,"year":9,"venue":66,"venueType":67,"publisher":68,"volumeIssuePages":69,"doi":70,"arxivId":71,"url":72,"firstPublicDate":73,"publicationStatus":16,"metadataStatus":74,"fulltextStatus":15,"era":10,"classicReason":75,"codeUrl":52,"cluster":11,"topics":76,"mdpi":77,"verification":78,"label":6,"fulltextRoute":79,"versionRead":80,"addedByCensus":81},"method",[62,63,64,65],"Weizhao Shao","Srinivasan Vijayarangan","Cong Li","George Kantor","2019 IEEE\u002FRSJ International Conference on Intelligent Robots and Systems (IROS)","conference","IEEE","pp. 370-377","10.1109\u002Firos40897.2019.8968012","1902.10741","https:\u002F\u002Fdoi.org\u002F10.1109\u002FIROS40897.2019.8968012","2019-02-27","metadata_verified","not_applicable",[11],false,"corrected","arXiv","arXiv v1 (2019-02-27) for full text and Table I; IEEE version of record (IROS 2019) HTML read through NTU access and matches arXiv v1 in text; VoR Table I image later viewed by the verifier (identical to arXiv v1)",true,[83,89,94,98,103,107,113],{"category":84,"model":85,"canonical":85,"role":86,"dataset":52,"specs":87,"locator":88},"stereo_camera","two megapixel cameras (model not reported)","method input","stereo pair on the custom platform","Sec. VIII-A; Fig. 1(a)",{"category":90,"model":91,"canonical":91,"role":86,"dataset":52,"specs":92,"locator":93},"lidar","16 scan-line LiDAR (model not reported)","3D LiDAR on the custom platform","Sec. VIII-A",{"category":95,"model":96,"canonical":96,"role":86,"dataset":52,"specs":97,"locator":93},"imu","IMU (model not reported)","400 Hz",{"category":99,"model":100,"canonical":100,"role":101,"dataset":52,"specs":102,"locator":93},"compute","4 GHz computer with 4 physical cores (model not reported)","compute for runtime","onboard computer of the custom platform",{"category":104,"model":105,"canonical":105,"role":86,"dataset":52,"specs":106,"locator":93},"other","custom microcontroller-based time synchronization circuit","synchronizes cameras, LiDAR, IMU and computer by simulating GPS time signals",{"category":108,"model":109,"canonical":109,"role":110,"dataset":52,"specs":111,"locator":112},"tls_scanner","Faro time-of-flight laser scanner (model not reported)","reference or ground truth","scans used as the reference model for mean registration error","Sec. I; Sec. VIII-B",{"category":99,"model":114,"canonical":114,"role":101,"dataset":115,"specs":116,"locator":117},"Intel i7-4790 desktop CPU","EuRoC MAV","3.60 GHz; EuRoC VIO results obtained in real time","Sec. VIII-C",[],{"totalRows":120,"groupCount":121,"groups":122,"others":235},12,2,[123,205],{"slug":124,"group":125,"sourceId":5,"sourceLabel":6,"table":126,"selfRows":127,"metrics":128,"seqs":137,"entrants":154,"cells":159,"outcomes":196,"locators":200,"hardware":201,"wordings":202,"notes":203},"vilslam2019-table-i","vilslam2019:Table I","Table I",10,[129,133],{"label":130,"unit":131,"statistic":132,"alignment":49},"FDE final drift error","%","not_reported",{"label":134,"unit":135,"statistic":136,"alignment":132},"MRE mean registration error to Faro scans","m","mean",[138,142,145,148,151],{"dataset":139,"sequence":140,"environment":141},"VIL-SLAM custom datasets","Highbay (total length 118, unit not printed)","indoor warehouse highbay, open, structured, feature-rich, frequent occlusions",{"dataset":139,"sequence":143,"environment":144},"Hallway (total length 103, unit not printed)","featureless hallway, LiDAR degenerate along traversal",{"dataset":139,"sequence":146,"environment":147},"Tunnel (total length 85, unit not printed)","tunnel, LiDAR degenerate along traversal",{"dataset":139,"sequence":149,"environment":150},"Huge Loop (total length 318, unit not printed)","hallway and highbay combined, ending by re-entering the highbay after a long narrow corridor",{"dataset":139,"sequence":152,"environment":153},"Outdoor (total length 528, unit not printed)","outdoor road with gentle slope, pedestrians and cars",[155,156],{"name":7,"methodId":5,"linkable":81,"proposed":81,"self":81},{"name":157,"methodId":158,"linkable":81,"proposed":77,"self":77},"LOAM","loam2014",[160,164,167,168,170,172,174,176,178,180,181,182,183,186,187,188,190,193,194,195],[161,161,161,162,163,161,163,163,161],0,0.08,-1,[165,161,161,166,163,161,163,163,161],1,0.56,[161,165,161,162,163,161,163,163,161],[165,165,161,169,163,161,163,163,161],0.22,[161,161,165,171,163,161,163,163,161],0.61,[165,161,165,173,163,161,163,163,161],0.91,[161,165,165,175,163,161,163,163,161],0.1,[165,165,165,177,163,161,163,163,161],0.27,[161,161,121,179,163,161,163,163,161],1.86,[165,161,121,52,161,161,163,163,161],[161,165,121,52,165,161,163,163,161],[165,165,121,52,165,161,163,163,161],[161,161,184,185,163,161,163,163,161],3,0.01,[165,161,184,52,161,161,163,163,161],[161,165,184,169,163,161,163,163,161],[165,165,184,189,121,161,163,163,161],0.36,[161,161,191,192,163,161,163,163,161],4,0.02,[165,161,191,192,163,161,163,163,161],[161,165,191,52,165,161,163,163,161],[165,165,191,52,165,161,163,163,161],[197,198,199],"failed (marked '-' not finished in Table I)","not_reported (marked 'x' missing data in Table I)","LOAM map evaluated only up to its failure point (Sec. VIII-B)",[126],[],[],[204],"Author-collected sequences that start and end at the same point; FDE = final drift error of LiDAR mapping odometry (no loop closure) as % of distance; MRE = mean distance from map points to the closest point of Faro reference scans after aligning the map; LOAM = laboshinl\u002Floam_velodyne implementation; '-' not finished, 'x' missing data",{"slug":206,"group":207,"sourceId":5,"sourceLabel":6,"table":208,"selfRows":121,"metrics":209,"seqs":214,"entrants":221,"cells":224,"outcomes":228,"locators":229,"hardware":231,"wordings":232,"notes":233},"vilslam2019-text-sec-viii-b","vilslam2019:Text Sec.VIII-B","Text Sec.VIII-B",[210,212],{"label":211,"unit":131,"statistic":132,"alignment":49},"FDE after loop closure (loop detected twice near the endpoint)",{"label":213,"unit":131,"statistic":132,"alignment":49},"FDE after loop closure (loop detected about 3 m from the end point)",[215,218],{"dataset":139,"sequence":216,"environment":217},"Hallway","featureless hallway",{"dataset":139,"sequence":219,"environment":220},"Tunnel","tunnel",[222],{"name":223,"methodId":5,"linkable":81,"proposed":81,"self":81},"VIL-SLAM (with loop closure)",[225,227],[161,161,161,226,163,161,163,163,161],0.05,[161,165,165,162,163,161,163,163,161],[],[230],"Sec. VIII-B",[],[],[234],"Final drift error after loop closure, stated in the text",[],1790510666129]