Main Content

다중 객체 추적기

다중 센서 다중 객체 추적기, 데이터 연결, 트랙 융합

다양한 센서의 정보를 융합하는 다중 객체 추적기를 만들 수 있습니다. 추적된 객체에 대해 단일 가설을 관리하려면 trackerGNN을 사용합니다. 추적된 객체에 대해 여러 가설을 관리하려면 trackerTOMHT를 사용합니다. 추적된 객체에 여러 가능한 탐지를 할당하려면 trackerJPDA를 사용합니다. PHD(확률 가설 밀도) 함수를 사용하여 추적된 객체를 나타내려면 trackerPHD를 사용합니다. 그리드 기반 점유 증거(occupancy evidence) 접근법을 사용하여 객체를 추적하려면 trackerGridRFS를 사용합니다. 추적 센서나 추적기가 생성한 트랙을 융합하고 분산화된 추적 시스템을 설계하려면 trackFuser를 사용합니다.

함수

모두 확장

assignauctionAssignment using auction global nearest neighbor
assignjvJonker-Volgenant global nearest neighbor assignment algorithm
assignkbestAssignment using k-best global nearest neighbor
assignkbestsdK-best S-D solution that minimizes total cost of assignment
assignmunkresMunkres global nearest neighbor assignment algorithm
assignsdS-D assignment using Lagrangian relaxation
assignTOMHTTrack-oriented multi-hypotheses tracking assignment
jpdaEventsFeasible joint events for trackerJPDA (R2019a 이후)
partitionDetectionsPartition detections based on distance (R2019a 이후)
mergeDetectionsMerge detections into clustered detections (R2021b 이후)
trackerGNNMulti-sensor, multi-object tracker using GNN assignment
trackerJPDAJoint probabilistic data association tracker (R2019a 이후)
trackerTOMHTMulti-hypothesis, multi-sensor, multi-object tracker
trackerPHDMulti-sensor, multi-object PHD tracker (R2019a 이후)
trackerGridRFSGrid-based multi-object tracker (R2020b 이후)
smootherJIPDAJoint probabilistic data association smoother (R2023a 이후)
dynamicEvidentialGridMapDynamic grid map output from trackerGridRFS (R2021a 이후)
objectDetectionReport for single object detection
objectDetectionDelaySimulate out-of-sequence object detections (R2022a 이후)
getTrackPositionsReturns updated track positions and position covariance matrix
getTrackVelocitiesObtain updated track velocities and velocity covariance matrix
clusterTrackBranchesCluster track-oriented multi-hypothesis history
compatibleTrackBranchesFormulate global hypotheses from clusters
pruneTrackBranchesPrune track branches with low likelihood
trackHistoryLogicConfirm and delete tracks based on recent track history
trackScoreLogicConfirm and delete tracks based on track score
trackBranchHistoryTrack-oriented MHT branching and branch history
trackingSensorConfiguration Represent sensor configuration for tracking (R2019a 이후)
trackFuserSingle-hypothesis track-to-track fuser (R2019b 이후)
trackingArchitectureTracking system-of-system architecture (R2021a 이후)
staticDetectionFuserStatic fusion of synchronous sensor detections
objectTrackSingle object track report (R2019b 이후)
fusecovintCovariance fusion using covariance intersection
fusecovunionCovariance fusion using covariance union
fusexcovCovariance fusion using cross-covariance
fuserSourceConfiguration Configuration of source used with track fuser (R2019b 이후)
triangulateLOSTriangulate multiple line-of-sight detections

블록

모두 확장

Global Nearest Neighbor Multi Object TrackerMulti-sensor, multi-object tracker using GNN assignment (R2019b 이후)
Joint Probabilistic Data Association Multi Object TrackerJoint probabilistic data association tracker (R2019b 이후)
Track-Oriented Multi-Hypothesis TrackerTrack-Oriented Multi-Hypothesis Tracker (R2020a 이후)
Probability Hypothesis Density (PHD) TrackerMulti-sensor, multi-object PHD tracker (R2021a 이후)
Grid-Based Multi Object TrackerGrid-based multi-object tracker using random finite set approach (R2021b 이후)
Track-To-Track FuserTrack-to-Track Fusion (R2021a 이후)
Detection ConcatenationCombine detection reports from different sensors (R2021a 이후)
Track ConcatenationConcatenate tracks (R2021a 이후)

도움말 항목