[{"data":1,"prerenderedAt":481},["ShallowReactive",2],{"method-kinematicicp2025":3},{"method":4,"reference":56,"equipment":82,"figures":117,"results":118},{"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":23,"limitations":27,"sensors":32,"platform":35,"estimator":37,"association":38,"timeModel":39,"deskew":40,"loopClosure":41,"globalOptimization":41,"mapRepresentation":42,"prior":43,"outputGeometry":44,"compute":45,"codeUrl":46,"codeLicense":47,"relatedVersions":48},"kinematicicp2025","Guadagnino et al., 2025b","Kinematic-ICP","Kinematic-ICP: Enhancing LiDAR Odometry with Kinematic Constraints for Wheeled Mobile Robots Moving on Planar Surfaces",2025,"recent","C05","odometry_with_local_mapping","Kinematic-ICP 針對在平面上移動、配備 3D LiDAR 的輪式機器人，把單輪車（unicycle）運動學模型放進點到點 ICP 最佳化，並以輪式里程計為初值與正則化項，使估計結果符合平台運動限制。正則化強度依情況自適應調整，以在特徵不足的走廊中更信任輪式里程計。作者回報已部署於 Dexory 倉儲機器人隊伍。","Adds unicycle kinematic constraints and adaptive wheel-odometry regularization to point-to-point ICP for wheeled robots on planar floors, improving robustness in feature-poor corridors.","full_text_reviewed","peer_reviewed_published","supplementary","在營運中倉庫（0.35 ha 至 9.45 ha）測試，屬既有建築室內；平面地板假設可能適用於樓板完成後的室內巡檢，但不適用於不平整工地地面（推論）。",[20,21,22],"completed_building","controlled_experiment","independent_reference",[24,25,26],"Better RPE and ATE than wheel odometry on all seven sequences and the lowest RPE of all methods on every sequence (Table II)","Authors' text claims better results than KISS-ICP on all sequences and consistent outperformance of EKF and Fuse, but Table II shows exceptions in ATE: Palace 1.56 m versus 0.69 m (KISS-ICP), 0.78 m (EKF) and 0.69 m (Fuse); WO + 2D KISS-ICP lower on Campus 1 (0.26 vs 0.42 m) and Warehouse Large (4.11 vs 4.42 m) (Sec. V-C; Table II)","Deployed on a commercial warehouse robot fleet (abstract; Sec. VI)",[28,29,30,31],"Planar-surface assumption: on uneven park terrain it performs slightly worse than KISS-ICP variants because slippage, rolling and pitching are not modelled (Sec. V-C)","Warehouse accuracy measured against Cartographer SLAM output rather than an independent reference (Sec. V-A)","Adaptive regularization is not always best: fixed beta = 0.01 gives lower Warehouse Small errors (RPE 0.39% vs 0.53%, ATE 0.20 vs 0.26 m) and larger fixed beta gives lower Palace ATE (Table III)","Requires the LiDAR-to-base extrinsic calibration and robot wheel odometry (Sec. III; Sec. V-A)",[33,34],"3D LiDAR (Robosense Bpearl, Hesai XT32)","wheel odometry",[36],"wheeled UGV","point-to-point ICP (KISS-ICP based) with unicycle kinematic model and adaptive regularization toward the wheel-odometry initial guess","point-to-point","discrete planar poses","de-skewing in preprocessing inherited from KISS-ICP (Sec. III); motion source for de-skewing not detailed in text read","none","voxel-grid local map as in KISS-ICP (author-stated, Sec. III)","wheel odometry; planar-surface assumption; LiDAR-to-base extrinsic required","planar odometry; export format not_reported","100 Hz on a single CPU core (authors' comparison with Fuse at ~10 Hz, Sec. V-C)","https:\u002F\u002Fgithub.com\u002FPRBonn\u002Fkinematic-icp","MIT (LICENSE file checked)",[49,53],{"relation":50,"title":51,"doi_or_url":52},"preprint","Kinematic-ICP (arXiv v3, accepted at ICRA 2025)","https:\u002F\u002Farxiv.org\u002Fabs\u002F2410.10277",{"relation":54,"title":55,"doi_or_url":46},"code_release","PRBonn\u002Fkinematic-icp",{"id":5,"kind":57,"shortName":7,"title":8,"authors":58,"year":9,"venue":67,"venueType":68,"publisher":69,"volumeIssuePages":70,"doi":71,"arxivId":72,"url":73,"firstPublicDate":74,"publicationStatus":16,"metadataStatus":75,"fulltextStatus":15,"era":10,"classicReason":76,"codeUrl":46,"cluster":11,"topics":77,"mdpi":78,"verification":79,"label":6,"fulltextRoute":80,"versionRead":81,"addedByCensus":78},"method",[59,60,61,62,63,64,65,66],"Tiziano Guadagnino","Benedikt Mersch","Ignacio Vizzo","Saurabh Gupta","Meher V.R. Malladi","Luca Lobefaro","Guillaume Doisy","Cyrill Stachniss","2025 IEEE International Conference on Robotics and Automation (ICRA)","conference","IEEE","pp. 11090-11096","10.1109\u002Ficra55743.2025.11128503","2410.10277","https:\u002F\u002Fapi.crossref.org\u002Fworks?query.bibliographic=Kinematic-ICP...","2024-10-14","metadata_verified","not_applicable",[11],false,"corrected","arXiv","arXiv v3 (2410.10277v3, 2025-02-10, 'Accepted at ICRA 2025'); IEEE ICRA 2025 version of record not opened",[83,90,95,100,103,106,112],{"category":84,"model":85,"canonical":85,"role":86,"dataset":87,"specs":88,"locator":89},"platform","Dexory robot","method input",null,"differential drive with front and back caster wheels; extendable 12 m tower; about 500 kg; provides wheel-encoder odometry","Sec. V-A-1; Sec. V-B",{"category":91,"model":92,"canonical":92,"role":86,"dataset":87,"specs":93,"locator":94},"lidar","Bpearl","90 x 360 deg hemispherical 32-beam LiDAR, 10 Hz","Sec. V-A-1",{"category":96,"model":97,"canonical":87,"role":86,"dataset":87,"specs":98,"locator":99},"wheel_or_leg_odometry","not_reported","wheel-encoder odometry of the Dexory robot and of the Husky, used as initial guess and regularization prior","Sec. III; Sec. V-A-1",{"category":84,"model":101,"canonical":101,"role":86,"dataset":87,"specs":102,"locator":94},"Clearpath Husky A200","four-wheeled skid-steering robot with wheel-encoder odometry",{"category":91,"model":104,"canonical":104,"role":86,"dataset":87,"specs":105,"locator":94},"Hesai LiDAR XT32","10 Hz",{"category":107,"model":108,"canonical":108,"role":109,"dataset":87,"specs":110,"locator":111},"total_station","Leica Nova MS60","reference or ground truth","tracks a reflective prism on the robot; angular accuracy 0.0003 deg, range accuracy 3 mm or better; initially time-synchronized with the robot","Sec. V-A-1; Fig. 2",{"category":113,"model":97,"canonical":87,"role":114,"dataset":87,"specs":115,"locator":116},"compute","compute for runtime","Kinematic-ICP runs at 100 Hz on a single CPU core; Fuse at about 10 Hz","Sec. V-C",[],{"totalRows":119,"groupCount":120,"groups":121,"others":480},19,3,[122,371,453],{"slug":123,"group":124,"sourceId":5,"sourceLabel":6,"table":125,"selfRows":126,"metrics":127,"seqs":136,"entrants":155,"cells":171,"outcomes":365,"locators":366,"hardware":367,"wordings":368,"notes":369},"kinematicicp2025-table-ii","kinematicicp2025:Table II","Table II",14,[128,132],{"label":129,"unit":130,"statistic":131,"alignment":97},"RPE [%] (KITTI metric, 1-100 m segments)","%","mean",{"label":133,"unit":134,"statistic":135,"alignment":97},"ATE [m] (root mean squared absolute translation error after alignment)","m","RMSE",[137,141,143,145,148,151,153],{"dataset":138,"sequence":139,"environment":140},"authors' warehouse and campus sequences","Campus 0","flat campus pavement, Clearpath Husky; reference Leica Nova MS60 total station",{"dataset":138,"sequence":142,"environment":140},"Campus 1",{"dataset":138,"sequence":144,"environment":140},"Campus 2",{"dataset":138,"sequence":146,"environment":147},"Palace","uneven park terrain (grass, curb walks) at Poppelsdorf Palace, Bonn; reference total station",{"dataset":138,"sequence":149,"environment":150},"Warehouse Small","operating warehouse with forklifts and people, Dexory robot; reference Cartographer SLAM output",{"dataset":138,"sequence":152,"environment":150},"Warehouse Mid",{"dataset":138,"sequence":154,"environment":150},"Warehouse Large",[156,158,162,164,166,168,170],{"name":157,"methodId":87,"linkable":78,"proposed":78,"self":78},"Wheel Odometry",{"name":159,"methodId":160,"linkable":161,"proposed":78,"self":78},"KISS-ICP [31]","kissicp2023",true,{"name":163,"methodId":87,"linkable":78,"proposed":78,"self":78},"WO + 3D KISS-ICP",{"name":165,"methodId":87,"linkable":78,"proposed":78,"self":78},"WO + 2D KISS-ICP",{"name":167,"methodId":87,"linkable":78,"proposed":78,"self":78},"EKF (robot_localization fusing WO + 2D KISS-ICP)",{"name":169,"methodId":87,"linkable":78,"proposed":78,"self":78},"Fuse (fixed-lag smoother fusing WO + 2D KISS-ICP)",{"name":7,"methodId":5,"linkable":161,"proposed":161,"self":161},[172,176,179,181,183,186,188,190,192,195,197,200,202,205,207,209,211,213,215,217,219,221,223,225,227,229,231,233,235,237,239,241,243,245,246,248,249,251,253,255,257,259,261,263,264,266,268,270,272,274,275,277,279,280,282,284,286,288,290,292,294,296,298,300,302,304,306,308,310,312,314,316,318,320,321,323,324,326,327,329,331,332,334,336,338,340,342,344,346,348,350,352,354,356,357,359,361,363],[173,173,173,174,175,173,175,175,173],0,4.93,-1,[173,177,173,178,175,173,175,175,173],1,3.25,[173,173,177,180,175,173,175,175,173],4.71,[173,177,177,182,175,173,175,175,173],6.52,[173,173,184,185,175,173,175,175,173],2,2.63,[173,177,184,187,175,173,175,175,173],1.87,[173,173,120,189,175,173,175,175,173],2.98,[173,177,120,191,175,173,175,175,173],4.66,[173,173,193,194,175,173,175,175,173],4,2.35,[173,177,193,196,175,173,175,175,173],1.74,[173,173,198,199,175,173,175,175,173],5,0.89,[173,177,198,201,175,173,175,175,173],16.62,[173,173,203,204,175,173,175,175,173],6,4.86,[173,177,203,206,175,173,175,175,173],108.11,[177,173,173,208,175,173,175,175,173],4.9,[177,177,173,210,175,173,175,175,173],0.31,[177,173,177,212,175,173,175,175,173],5.37,[177,177,177,214,175,173,175,175,173],0.5,[177,173,184,216,175,173,175,175,173],8.82,[177,177,184,218,175,173,175,175,173],0.29,[177,173,120,220,175,173,175,175,173],3.75,[177,177,120,222,175,173,175,175,173],0.69,[177,173,193,224,175,173,175,175,173],59.7,[177,177,193,226,175,173,175,175,173],7.36,[177,173,198,228,175,173,175,175,173],18.99,[177,177,198,230,175,173,175,175,173],54.7,[177,173,203,232,175,173,175,175,173],164.78,[177,177,203,234,175,173,175,175,173],23.5,[184,173,173,236,175,173,175,175,173],4.64,[184,177,173,238,175,173,175,175,173],0.34,[184,173,177,240,175,173,175,175,173],3.82,[184,177,177,242,175,173,175,175,173],0.32,[184,173,184,244,175,173,175,175,173],5.97,[184,177,184,210,175,173,175,175,173],[184,173,120,247,175,173,175,175,173],4.14,[184,177,120,218,175,173,175,175,173],[184,173,193,250,175,173,175,175,173],5.32,[184,177,193,252,175,173,175,175,173],5.4,[184,173,198,254,175,173,175,175,173],5.85,[184,177,198,256,175,173,175,175,173],28.71,[184,173,203,258,175,173,175,175,173],1.48,[184,177,203,260,175,173,175,175,173],17.52,[120,173,173,262,175,173,175,175,173],4.43,[120,177,173,242,175,173,175,175,173],[120,173,177,265,175,173,175,175,173],3.94,[120,177,177,267,175,173,175,175,173],0.26,[120,173,184,269,175,173,175,175,173],6.77,[120,177,184,271,175,173,175,175,173],0.25,[120,173,120,273,175,173,175,175,173],4.63,[120,177,120,218,175,173,175,175,173],[120,173,193,276,175,173,175,175,173],2.02,[120,177,193,278,175,173,175,175,173],1.01,[120,173,198,278,175,173,175,175,173],[120,177,198,281,175,173,175,175,173],6.48,[120,173,203,283,175,173,175,175,173],1.13,[120,177,203,285,175,173,175,175,173],4.11,[193,173,173,287,175,173,175,175,173],6.28,[193,177,173,289,175,173,175,175,173],0.47,[193,173,177,291,175,173,175,175,173],5.84,[193,177,177,293,175,173,175,175,173],1.61,[193,173,184,295,175,173,175,175,173],6.13,[193,177,184,297,175,173,175,175,173],0.38,[193,173,120,299,175,173,175,175,173],4.98,[193,177,120,301,175,173,175,175,173],0.78,[193,173,193,303,175,173,175,175,173],0.87,[193,177,193,305,175,173,175,175,173],0.86,[193,173,198,307,175,173,175,175,173],0.9,[193,177,198,309,175,173,175,175,173],9.53,[193,173,203,311,175,173,175,175,173],0.85,[193,177,203,313,175,173,175,175,173],13.85,[198,173,173,315,175,173,175,175,173],4.16,[198,177,173,317,175,173,175,175,173],0.3,[198,173,177,319,175,173,175,175,173],6.69,[198,177,177,311,175,173,175,175,173],[198,173,184,322,175,173,175,175,173],3.92,[198,177,184,317,175,173,175,175,173],[198,173,120,325,175,173,175,175,173],3.23,[198,177,120,222,175,173,175,175,173],[198,173,193,328,175,173,175,175,173],0.61,[198,177,193,330,175,173,175,175,173],0.4,[198,173,198,222,175,173,175,175,173],[198,177,198,333,175,173,175,175,173],9.18,[198,173,203,335,175,173,175,175,173],1.36,[198,177,203,337,175,173,175,175,173],4.99,[203,173,173,339,175,173,175,175,173],2.97,[203,177,173,341,175,173,175,175,173],0.28,[203,173,177,343,175,173,175,175,173],2.93,[203,177,177,345,175,173,175,175,173],0.42,[203,173,184,347,175,173,175,175,173],2.13,[203,177,184,349,175,173,175,175,173],0.22,[203,173,120,351,175,173,175,175,173],2.38,[203,177,120,353,175,173,175,175,173],1.56,[203,173,193,355,175,173,175,175,173],0.53,[203,177,193,267,175,173,175,175,173],[203,173,198,358,175,173,175,175,173],0.46,[203,177,198,360,175,173,175,175,173],4.44,[203,173,203,362,175,173,175,175,173],0.39,[203,177,203,364,175,173,175,175,173],4.42,[],[125],[],[],[370],"RPE is the KITTI average translation error over 1, 2, 5, 10, 20, 50 and 100 m segments (%); ATE is RMS absolute translation error after alignment (m); warehouse reference is Cartographer, campus and park reference is the total station",{"slug":372,"group":373,"sourceId":5,"sourceLabel":6,"table":374,"selfRows":193,"metrics":375,"seqs":380,"entrants":383,"cells":398,"outcomes":447,"locators":448,"hardware":449,"wordings":450,"notes":451},"kinematicicp2025-table-iii","kinematicicp2025:Table III","Table III",[376,378],{"label":377,"unit":130,"statistic":131,"alignment":97},"RPE [%]",{"label":379,"unit":134,"statistic":135,"alignment":97},"ATE [m]",[381,382],{"dataset":138,"sequence":146,"environment":147},{"dataset":138,"sequence":149,"environment":150},[384,386,388,390,392,394,396],{"name":385,"methodId":87,"linkable":78,"proposed":78,"self":78},"Fixed beta = 0.01",{"name":387,"methodId":87,"linkable":78,"proposed":78,"self":78},"Fixed beta = 0.1",{"name":389,"methodId":87,"linkable":78,"proposed":78,"self":78},"Fixed beta = 1.0",{"name":391,"methodId":87,"linkable":78,"proposed":78,"self":78},"Fixed beta = 10.0",{"name":393,"methodId":87,"linkable":78,"proposed":78,"self":78},"Fixed beta = 100.0",{"name":395,"methodId":87,"linkable":78,"proposed":78,"self":78},"No Regularization",{"name":397,"methodId":5,"linkable":161,"proposed":161,"self":161},"Kinematic-ICP (adaptive regularization)",[399,401,403,404,406,408,410,412,414,416,418,419,421,423,425,427,429,431,432,434,436,438,439,441,443,444,445,446],[173,173,173,400,175,173,175,175,173],2.39,[173,177,173,402,175,173,175,175,173],1.7,[173,173,177,362,175,173,175,175,173],[173,177,177,405,175,173,175,175,173],0.2,[177,173,173,407,175,173,175,175,173],1.79,[177,177,173,409,175,173,175,175,173],1.41,[177,173,177,411,175,173,175,175,173],1.81,[177,177,177,413,175,173,175,175,173],0.59,[184,173,173,415,175,173,175,175,173],2.56,[184,177,173,417,175,173,175,175,173],0.44,[184,173,177,120,175,173,175,175,173],[184,177,177,420,175,173,175,175,173],0.62,[120,173,173,422,175,173,175,175,173],3.71,[120,177,173,424,175,173,175,175,173],0.33,[120,173,177,426,175,173,175,175,173],3.03,[120,177,177,428,175,173,175,175,173],1.75,[193,173,173,430,175,173,175,175,173],3.99,[193,177,173,210,175,173,175,175,173],[193,173,177,433,175,173,175,175,173],4.36,[193,177,177,435,175,173,175,175,173],1.3,[198,173,173,437,175,173,175,175,173],3.72,[198,177,173,238,175,173,175,175,173],[198,173,177,440,175,173,175,175,173],3.77,[198,177,177,442,175,173,175,175,173],1.59,[203,173,173,351,175,173,175,175,173],[203,177,173,353,175,173,175,175,173],[203,173,177,355,175,173,175,175,173],[203,177,177,267,175,173,175,175,173],[],[374],[],[],[452],"Ablation on regularization of the wheel-odometry translation prior; all rows are Kinematic-ICP variants",{"slug":454,"group":455,"sourceId":5,"sourceLabel":6,"table":456,"selfRows":177,"metrics":457,"seqs":461,"entrants":464,"cells":468,"outcomes":473,"locators":474,"hardware":475,"wordings":477,"notes":478},"kinematicicp2025-text-sec-v-c","kinematicicp2025:Text Sec. V-C","Text Sec. V-C",[458],{"label":459,"unit":460,"statistic":97,"alignment":41},"runs at ... Hz","Hz",[462],{"dataset":463,"sequence":97,"environment":97},"authors' sequences",[465,466],{"name":7,"methodId":5,"linkable":161,"proposed":161,"self":161},{"name":467,"methodId":87,"linkable":78,"proposed":78,"self":78},"Fuse",[469,471],[173,173,173,470,175,173,173,175,173],100,[177,173,173,472,175,173,175,175,173],10,[],[116],[476],"single-core CPU (model not reported)",[],[479],"Runtime comparison stated in text; Fuse value given as approximately 10 Hz",[],1790510661538]