[{"data":1,"prerenderedAt":564},["ShallowReactive",2],{"method-fastlivo2rc2025":3},{"method":4,"reference":54,"equipment":77,"figures":128,"results":129},{"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":28,"sensors":31,"platform":35,"estimator":38,"association":39,"timeModel":40,"deskew":41,"loopClosure":42,"globalOptimization":43,"mapRepresentation":44,"prior":43,"outputGeometry":45,"compute":46,"codeUrl":47,"codeLicense":48,"relatedVersions":49},"fastlivo2rc2025","Zhou et al., 2025","FAST-LIVO2 on Resource-Constrained Platforms","FAST-LIVO2 on Resource-Constrained Platforms: LiDAR-Inertial-Visual Odometry With Efficient Memory and Computation",2025,"recent","C07","odometry_with_local_mapping","此研究針對邊緣運算平台精簡 FAST-LIVO2：以光達退化評估決定何時需要影像更新，在光達約束充足時減少視覺幀，降低計算量；地圖改為小範圍的統一視覺光達局部地圖加上稀疏的長期視覺地圖，以限制記憶體。作者在 Hilti（資料集描述含工地序列）16 個序列與私人序列（礦坑隧道、黑暗樹林、野外公園等）上測試，並在約 100 美元的 RK3588 ARM 板上於地下停車場與夜間街道進行即時定位測試。","A lightweight FAST-LIVO2 variant with degeneration-aware adaptive visual frame selection and a compact local plus long-term visual map for ARM edge devices.","full_text_reviewed","peer_reviewed_published","supplementary","作者以 Hilti 2022 與 2023 共 16 個序列評估軌跡精度，論文描述其場景含營建工地、辦公室與地下室，並使用 Hilti 官方評估工具。Hilti 2022 三個工地序列的 RMSE 為 Construction Ground 0.010 m、Construction Multilevel 0.023 m、Construction Stairs 0.170 m，其中樓梯序列明顯遜於 FAST-LIVO2 的 0.016 m。另在礦坑隧道與地下停車場做定性測試。證據屬軌跡層級，未報告點雲幾何精度。",[20,21,22,23],"public_benchmark","real_construction_site","underground_or_tunnel","independent_reference",[25,26,27],"On Hilti data, 33% lower per-frame runtime and 47% lower memory than FAST-LIVO2 with about 3 cm higher RMSE (abstract)","Return-to-origin drift below 2 cm on private sequences including a mining tunnel (Sec. V-C1)","37 ms per frame on RK3588 ARM (Sec. V-C2)",[29,30],"Accuracy slightly lower than FAST-LIVO2 on Hilti (average 0.063 m vs 0.034 m) (abstract","Table I) | Much larger error than FAST-LIVO2 on Construction Stairs (0.170 vs 0.016 m), Cupola (0.220 vs 0.121 m) and Attic to Upper Gallery (0.180 vs 0.069 m) (Table I) | Slightly worse than FAST-LIO2 in visually challenging sequences such as overexposed Outside Building and dark Large Room (Sec. V-B1) | Long-term visual map adds memory overhead (Sec. V-E2)",[32,33,34],"3D LiDAR (Hilti dataset LiDARs","Livox Mid-360 on the authors' rig","a small-FoV AVIA LiDAR, written 'Aivia' in Sec. V-D2, in the degeneration-detection test of Fig. 9, sequence not named) | IMU | camera (B\u002FW fisheye on the authors' rig) | 15 W onboard illuminator for extremely dark scenes",[36,37],"handheld | ground robot (Hilti robot-mounted sequences","type not_verified) | aerial dataset (MARS-LVIG HKIsland03 map ablation) | ARM real-time tests in an underground parking lot and a nighttime street (carrying mode not stated)","ESIKF with sequential updates (from FAST-LIVO2) plus a LiDAR-degeneration-aware adaptive visual frame selector","LiDAR point-to-plane and patch photometric errors as in FAST-LIVO2; images used only when LiDAR constraints are weak or keyframe criteria met","discrete poses; scan recombination (Sec. III)","Uses undistorted points of recombined scans (scan recombination); the undistortion step itself is inherited from FAST-LIVO2 and not re-described","none reported","none","Compact unified local visual-LiDAR voxel map (hash of 0.5 m root voxels with three-level octrees; typical edge 200 m, slid every 20 m of motion) plus a sparse long-term visual map (edge 800 m, slid every 100 m) that keeps visual points leaving the local map","sharp point cloud maps and colored point clouds (Sec. V-C1; Fig. 10)","Hilti mean per-frame time 25.99 ms on the x86 laptop (13th Gen Intel Core i9-13900HX) versus 35.66 ms for FAST-LIVO2, and 57.82 ms on the RK3588 board (4x Cortex-A76 + 4x Cortex-A55, up to 2.4 GHz, about 100 USD, CPU only) versus 75.87 ms; about 37 ms per frame in the onboard ARM tests; mean memory 1.7 GB versus 2.5 GB on Hilti",null,"not_verified",[50],{"relation":51,"title":52,"doi_or_url":53},"preprint","arXiv v1 (2025-01-23)","https:\u002F\u002Farxiv.org\u002Fabs\u002F2501.13876",{"id":5,"kind":55,"shortName":7,"title":8,"authors":56,"year":9,"venue":63,"venueType":64,"publisher":65,"volumeIssuePages":66,"doi":67,"arxivId":68,"url":53,"firstPublicDate":69,"publicationStatus":16,"metadataStatus":70,"fulltextStatus":15,"era":10,"classicReason":71,"codeUrl":47,"cluster":11,"topics":72,"mdpi":73,"verification":74,"label":6,"fulltextRoute":75,"versionRead":76,"addedByCensus":73},"method",[57,58,59,60,61,62],"Bingyang Zhou","Chunran Zheng","Ziming Wang","Fangcheng Zhu","Yixi Cai","Fu Zhang","IEEE Robotics and Automation Letters","journal","IEEE","10(8): 7931-7938","10.1109\u002Flra.2025.3581125","2501.13876","2025-01-23","metadata_verified","not_applicable",[11],false,"corrected","arXiv","arXiv 2501.13876v1 HTML (2025-01-23); IEEE RA-L version of record not compared",[78,86,91,97,100,106,111,118,124],{"category":79,"model":80,"canonical":81,"role":82,"dataset":83,"specs":84,"locator":85},"lidar","Livox Mid-360","Livox MID-360","method input","private dataset","not_reported","Sec. V-A2; Fig. 5",{"category":87,"model":88,"canonical":88,"role":82,"dataset":83,"specs":89,"locator":90},"camera","B\u002FW camera with a fisheye lens","equidistant projection model","Sec. V-A2, V-A3",{"category":92,"model":93,"canonical":93,"role":94,"dataset":83,"specs":95,"locator":96},"other","STM32 microcontroller","dataset sensor","hardware synchronization of LiDAR and camera","Sec. V-A2",{"category":92,"model":98,"canonical":98,"role":94,"dataset":83,"specs":99,"locator":96},"onboard illuminator","15 W, used in extremely dark environments",{"category":79,"model":101,"canonical":102,"role":94,"dataset":103,"specs":104,"locator":105},"AVIA (written 'Aivia' in Sec. V-D2)","Livox Avia","unnamed LiDAR-degeneration test sequence of Fig. 9 (dataset not stated)","small-FoV LiDAR, facing a wall in the degeneration test","Sec. IV-A2; Sec. V-D2; Fig. 9",{"category":92,"model":107,"canonical":107,"role":94,"dataset":108,"specs":109,"locator":110},"Hilti handheld and robot-mounted rigs (LiDAR, cameras, IMUs; models not stated in this paper)","Hilti'22 and Hilti'23","sensors at different frequencies","Sec. V-A1",{"category":112,"model":113,"canonical":113,"role":114,"dataset":115,"specs":116,"locator":117},"gnss","RTK","reference or ground truth","MARS-LVIG HKIsland03","RTK trajectory used as ground truth","Sec. V-E1",{"category":119,"model":120,"canonical":120,"role":121,"dataset":47,"specs":122,"locator":123},"compute","13th Gen Intel Core i9-13900HX (personal laptop)","compute for runtime","x86 platform","Sec. V-A3",{"category":119,"model":125,"canonical":125,"role":121,"dataset":47,"specs":126,"locator":127},"RK3588","octa-core 4x Cortex-A76 + 4x Cortex-A55, max 2.4 GHz, about 100 USD; CPU only","Sec. V-A3; Fig. 1",[],{"totalRows":130,"groupCount":131,"groups":132,"others":552},30,6,[133,444,489,528],{"slug":134,"group":135,"sourceId":5,"sourceLabel":6,"table":136,"selfRows":137,"metrics":138,"seqs":145,"entrants":196,"cells":217,"outcomes":436,"locators":438,"hardware":439,"wordings":440,"notes":441},"fastlivo2rc2025-table-i","fastlivo2rc2025:Table I","Table I",17,[139,143],{"label":140,"unit":141,"statistic":142,"alignment":84},"ATE (RMSE)","m","RMSE",{"label":144,"unit":141,"statistic":142,"alignment":84},"ATE (RMSE), Average",[146,150,152,155,158,161,164,167,170,174,176,178,181,184,187,190,193],{"dataset":147,"sequence":148,"environment":149},"Hilti'22","Construction Ground","construction site",{"dataset":147,"sequence":151,"environment":149},"Construction Multilevel",{"dataset":147,"sequence":153,"environment":154},"Construction Stairs","construction site, stairs",{"dataset":147,"sequence":156,"environment":157},"Long Corridor","long corridor",{"dataset":147,"sequence":159,"environment":160},"Cupola","indoor building (cupola)",{"dataset":147,"sequence":162,"environment":163},"Lower Gallery","indoor building (gallery)",{"dataset":147,"sequence":165,"environment":166},"Attic to Upper Gallery","indoor building (attic and gallery)",{"dataset":147,"sequence":168,"environment":169},"Outside Building","outdoor around building",{"dataset":171,"sequence":172,"environment":173},"Hilti'23","Floor 0","indoor floor",{"dataset":171,"sequence":175,"environment":173},"Floor 1",{"dataset":171,"sequence":177,"environment":173},"Floor 2",{"dataset":171,"sequence":179,"environment":180},"Basement","basement",{"dataset":171,"sequence":182,"environment":183},"Stairs","stairs",{"dataset":171,"sequence":185,"environment":186},"Parking 3x floors down","underground parking",{"dataset":171,"sequence":188,"environment":189},"Large room","large room",{"dataset":171,"sequence":191,"environment":192},"Large room (dark)","large room, dark",{"dataset":108,"sequence":194,"environment":195},"Average (16 sequences)","construction sites, offices, basements",[197,200,203,206,209,212,214],{"name":198,"methodId":5,"linkable":199,"proposed":199,"self":199},"Ours",true,{"name":201,"methodId":202,"linkable":199,"proposed":73,"self":73},"FAST-LIVO2","fastlivo2_2025",{"name":204,"methodId":205,"linkable":199,"proposed":73,"self":73},"FAST-LIO2","fastlio2_2022",{"name":207,"methodId":208,"linkable":199,"proposed":73,"self":73},"FAST-LIVO","fastlivo2022",{"name":210,"methodId":211,"linkable":199,"proposed":73,"self":73},"R3LIVE","r3live2022",{"name":213,"methodId":47,"linkable":73,"proposed":73,"self":73},"SDV-LOAM",{"name":215,"methodId":216,"linkable":199,"proposed":73,"self":73},"LVI-SAM","lvisam2021",[218,222,224,227,230,233,236,237,239,241,243,245,247,249,250,252,254,256,258,260,262,264,266,268,270,272,274,276,278,280,282,284,286,288,290,291,293,295,296,297,299,301,303,305,307,309,311,313,315,316,319,321,323,324,326,328,330,333,334,336,337,338,340,341,343,344,345,346,347,349,351,354,355,357,359,361,363,364,366,367,368,369,370,372,373,376,377,378,380,382,384,386,389,390,391,393,395,397,398,400,402,403,404,405,407,409,411,412,414,416,418,420,421,424,426,428,430,432,434],[219,219,219,220,221,219,221,221,219],0,0.01,-1,[223,219,219,220,221,219,221,221,219],1,[225,219,219,226,221,219,221,221,219],2,0.013,[228,219,219,229,221,219,221,221,219],3,0.022,[231,219,219,232,221,219,221,221,219],4,0.021,[234,219,219,235,221,219,221,221,219],5,25.121,[131,219,219,47,219,219,221,221,219],[219,219,223,238,221,219,221,221,219],0.023,[223,219,223,240,221,219,221,221,219],0.02,[225,219,223,242,221,219,221,221,219],0.044,[228,219,223,244,221,219,221,221,219],0.052,[231,219,223,246,221,219,221,221,219],0.024,[234,219,223,248,221,219,221,221,219],12.561,[131,219,223,47,219,219,221,221,219],[219,219,225,251,221,219,221,221,219],0.17,[223,219,225,253,221,219,221,221,219],0.016,[225,219,225,255,221,219,221,221,219],0.32,[228,219,225,257,221,219,221,221,219],0.241,[231,219,225,259,221,219,221,221,219],0.784,[234,219,225,261,221,219,221,221,219],9.212,[131,219,225,263,221,219,221,221,219],9.142,[219,219,228,265,221,219,221,221,219],0.054,[223,219,228,267,221,219,221,221,219],0.067,[225,219,228,269,221,219,221,221,219],0.064,[228,219,228,271,221,219,221,221,219],0.065,[231,219,228,273,221,219,221,221,219],0.061,[234,219,228,275,221,219,221,221,219],19.531,[131,219,228,277,221,219,221,221,219],6.312,[219,219,231,279,221,219,221,221,219],0.22,[223,219,231,281,221,219,221,221,219],0.121,[225,219,231,283,221,219,221,221,219],0.25,[228,219,231,285,221,219,221,221,219],0.182,[231,219,231,287,221,219,221,221,219],2.142,[234,219,231,289,221,219,221,221,219],9.321,[131,219,231,47,219,219,221,221,219],[219,219,234,292,221,219,221,221,219],0.018,[223,219,234,294,221,219,221,221,219],0.007,[225,219,234,246,221,219,221,221,219],[228,219,234,229,221,219,221,221,219],[231,219,234,298,221,219,221,221,219],0.008,[234,219,234,300,221,219,221,221,219],11.232,[131,219,234,302,221,219,221,221,219],2.281,[219,219,131,304,221,219,221,221,219],0.18,[223,219,131,306,221,219,221,221,219],0.069,[225,219,131,308,221,219,221,221,219],0.72,[228,219,131,310,221,219,221,221,219],0.621,[231,219,131,312,221,219,221,221,219],2.412,[234,219,131,314,221,219,221,221,219],4.551,[131,219,131,47,219,219,221,221,219],[219,219,317,318,221,219,221,221,219],7,0.041,[223,219,317,320,221,219,221,221,219],0.035,[225,219,317,322,221,219,221,221,219],0.028,[228,219,317,244,221,219,221,221,219],[231,219,317,325,221,219,221,221,219],0.029,[234,219,317,327,221,219,221,221,219],2.622,[131,219,317,329,221,219,221,221,219],0.952,[219,219,331,332,221,219,221,221,219],8,0.032,[223,219,331,232,221,219,221,221,219],[225,219,331,335,221,219,221,221,219],0.031,[228,219,331,229,221,219,221,221,219],[231,219,331,246,221,219,221,221,219],[234,219,331,339,221,219,221,221,219],4.621,[131,219,331,47,219,219,221,221,219],[219,219,342,292,221,219,221,221,219],9,[223,219,342,238,221,219,221,221,219],[225,219,342,335,221,219,221,221,219],[228,219,342,229,221,219,221,221,219],[231,219,342,246,221,219,221,221,219],[234,219,342,348,221,219,221,221,219],7.951,[131,219,342,350,221,219,221,221,219],8.682,[219,219,352,353,221,219,221,221,219],10,0.038,[223,219,352,229,221,219,221,221,219],[225,219,352,356,221,219,221,221,219],0.083,[228,219,352,358,221,219,221,221,219],0.048,[231,219,352,360,221,219,221,221,219],0.046,[234,219,352,362,221,219,221,221,219],7.912,[131,219,352,47,219,219,221,221,219],[219,219,365,246,221,219,221,221,219],11,[223,219,365,253,221,219,221,221,219],[225,219,365,353,221,219,221,221,219],[228,219,365,320,221,219,221,221,219],[231,219,365,246,221,219,221,221,219],[234,219,365,371,221,219,221,221,219],6.151,[131,219,365,47,219,219,221,221,219],[219,219,374,375,221,219,221,221,219],12,0.012,[223,219,374,292,221,219,221,221,219],[225,219,374,251,221,219,221,221,219],[228,219,374,379,221,219,221,221,219],0.152,[231,219,374,381,221,219,221,221,219],0.11,[234,219,374,383,221,219,221,221,219],9.032,[131,219,374,385,221,219,221,221,219],3.584,[219,219,387,388,221,219,221,221,219],13,0.095,[223,219,387,332,221,219,221,221,219],[225,219,387,255,221,219,221,221,219],[228,219,387,392,221,219,221,221,219],0.356,[231,219,387,394,221,219,221,221,219],0.462,[234,219,387,396,221,219,221,221,219],19.952,[131,219,387,47,219,219,221,221,219],[219,219,399,322,221,219,221,221,219],14,[223,219,399,401,221,219,221,221,219],0.026,[225,219,399,322,221,219,221,221,219],[228,219,399,335,221,219,221,221,219],[231,219,399,320,221,219,221,221,219],[234,219,399,406,221,219,221,221,219],16.781,[131,219,399,408,221,219,221,221,219],0.563,[219,219,410,242,221,219,221,221,219],15,[223,219,410,360,221,219,221,221,219],[225,219,410,413,221,219,221,221,219],0.04,[228,219,410,415,221,219,221,221,219],0.053,[231,219,410,417,221,219,221,221,219],0.059,[234,219,410,419,221,219,221,221,219],15.012,[131,219,410,47,219,219,221,221,219],[219,223,422,423,221,219,221,221,223],16,0.063,[223,223,422,425,221,219,221,221,223],0.034,[225,223,422,427,221,219,221,221,223],0.138,[228,223,422,429,221,219,221,221,223],0.123,[231,223,422,431,221,219,221,221,223],0.391,[234,223,422,433,221,219,221,221,223],11.347,[131,223,422,435,221,219,221,221,223],4.502,[437],"failed (x)",[136],[],[],[442,443],"ATE RMSE on 16 Hilti'22 and Hilti'23 sequences computed with the official Hilti evaluation tools; parameters of all methods tuned by the authors; x = system totally failed","ATE RMSE on 16 Hilti'22 and Hilti'23 sequences computed with the official Hilti evaluation tools; parameters of all methods tuned by the authors; x = system totally failed; Average row over the 16 sequences (handling of failures not stated)",{"slug":445,"group":446,"sourceId":5,"sourceLabel":6,"table":447,"selfRows":131,"metrics":448,"seqs":463,"entrants":466,"cells":468,"outcomes":481,"locators":482,"hardware":483,"wordings":486,"notes":487},"fastlivo2rc2025-table-ii","fastlivo2rc2025:Table II","Table II",[449,453,455,457,459,461],{"label":450,"unit":451,"statistic":452,"alignment":71},"LiDAR Part time, mean (standard error 5.35 ms)","ms","mean",{"label":454,"unit":451,"statistic":452,"alignment":71},"Visual Part time, mean (standard error 1.8 ms)",{"label":456,"unit":451,"statistic":452,"alignment":71},"Total time, mean (standard error 6.75 ms)",{"label":458,"unit":451,"statistic":452,"alignment":71},"LiDAR Part time, mean (standard error 2.36 ms)",{"label":460,"unit":451,"statistic":452,"alignment":71},"Visual Part time, mean (standard error 1.12 ms)",{"label":462,"unit":451,"statistic":452,"alignment":71},"Total time, mean (standard error 3.22 ms)",[464],{"dataset":108,"sequence":465,"environment":195},"mean over 16 sequences",[467],{"name":198,"methodId":5,"linkable":199,"proposed":199,"self":199},[469,471,473,475,477,479],[219,219,219,470,221,219,219,221,219],53.83,[219,223,219,472,221,219,219,221,219],3.99,[219,225,219,474,221,219,219,221,219],57.82,[219,228,219,476,221,219,223,221,219],23.36,[219,231,219,478,221,219,223,221,219],2.64,[219,234,219,480,221,219,223,221,219],25.99,[],[447],[484,485],"RK3588 (4x Cortex-A76 + 4x Cortex-A55, up to 2.4 GHz), CPU only","laptop, 13th Gen Intel Core i9-13900HX",[],[488],"Mean per-frame time over the 16 Hilti sequences, standard error given in metric_as_written; ARM runtime measured on CPU only",{"slug":490,"group":491,"sourceId":5,"sourceLabel":6,"table":492,"selfRows":231,"metrics":493,"seqs":497,"entrants":503,"cells":506,"outcomes":522,"locators":523,"hardware":524,"wordings":525,"notes":526},"fastlivo2rc2025-table-iii","fastlivo2rc2025:Table III","Table III",[494],{"label":495,"unit":496,"statistic":84,"alignment":71},"memory usage","GB",[498,499,500,501],{"dataset":147,"sequence":148,"environment":149},{"dataset":147,"sequence":151,"environment":149},{"dataset":147,"sequence":153,"environment":149},{"dataset":108,"sequence":194,"environment":502},"mixed",[504,505],{"name":198,"methodId":5,"linkable":199,"proposed":199,"self":199},{"name":201,"methodId":202,"linkable":199,"proposed":73,"self":73},[507,508,510,512,514,516,518,520],[219,219,219,228,221,219,219,221,219],[223,219,219,509,221,219,219,221,219],4.1,[219,219,223,511,221,219,219,221,219],3.7,[223,219,223,513,221,219,219,221,219],4.9,[219,219,225,515,221,219,219,221,219],1.3,[223,219,225,517,221,219,219,221,219],2.4,[219,219,228,519,221,219,219,221,219],1.7,[223,219,228,521,221,219,219,221,219],2.5,[],[492],[485],[],[527],"Memory usage on the x86 laptop; only the three Hilti'22 construction sequences and the Average row kept for the row cap",{"slug":529,"group":530,"sourceId":5,"sourceLabel":6,"table":531,"selfRows":223,"metrics":532,"seqs":535,"entrants":540,"cells":542,"outcomes":544,"locators":546,"hardware":548,"wordings":549,"notes":550},"fastlivo2rc2025-text-sec-v-c1","fastlivo2rc2025:Text Sec.V-C1","Text Sec.V-C1",[533],{"label":534,"unit":141,"statistic":84,"alignment":43},"return-to-origin drift",[536],{"dataset":537,"sequence":538,"environment":539},"private sequences","Mining Tunnel, Dark Woods, Wild Park, HIT Graffiti Wall, HW Corridor and others","mining tunnel, dark woods, park, corridors",[541],{"name":198,"methodId":5,"linkable":199,"proposed":199,"self":199},[543],[219,219,219,240,219,219,221,221,219],[545],"reported as less than 2 cm (upper bound)",[547],"Sec. V-C1; Fig. 6",[],[],[551],"Return-to-origin drift on private sequences that physically return to the start; value is an upper bound",[553,558],{"group":554,"slug":555,"sourceLabel":6,"table":556,"selfRows":223,"datasets":557},"fastlivo2rc2025:Text Sec.V-C2","fastlivo2rc2025-text-sec-v-c2","Text Sec.V-C2",[537],{"group":559,"slug":560,"sourceLabel":6,"table":561,"selfRows":223,"datasets":562},"fastlivo2rc2025:Text Sec.V-E1","fastlivo2rc2025-text-sec-v-e1","Text Sec.V-E1",[563],"MARS-LVIG",1790510657546]