Abstract Present Research paper suggests combined Bayesian framework for attribute-based simultaneous localization and map building (SLAM) in common instance of indefinite feature count and relation of data. Through modeling the evaluations and map of feature as RFSsrandom-finite-sets, the formulation is on SLAM attribute–based issue will be shown which combined predicts the count and features location and also trajectory of vehicle. More precisely, joint posterior dissemination of group-valued map & vehicle path will be propagated towards timely arrival of measurements, hence including both the management of feature and association of data into 1 recursion. And the objective of the thesis is applying & experimentally evaluating the RFSs manifold SLAM Vehicle in the 3D. And both real life and simulated datasets are verified for offering the accurate analysis and entire execution of suggested implementation. Moreover, to solve the consistent connection topic of transferring the objects, covariance region connection belief assignment will be implemented for evaluation of motion state & corresponding evidences like vision features & kinematics are synergized jointly to improve the efficiency of association through the verification fusion model. And concept proof with simulations is fruitfully analyzed & demonstrated.