Research On Simultaneous Localization And Map Building Technology Based On Information Filter

The simultaneous localization and map building (SLAM) is the key research of mobile robot navigation area. In the SLAM problem the robot starts in an unknown location in an unknown environment and then incrementally builds a navigation map of the environment while simultaneously uses this map to compute its location. A solution to the SLAM problem has the extremely important theory and application value. The possession of the SLAM capability is considered to be the key to make a robot truly autonomous.The EKF based estimation algorithm is the basic solution to the SLAM problem. It can provide the optimal solution to the SLAM problem, but has the problem of large computational complexity, which is a quadratic relation with environment features. This problem limits its application in large-scale environment. This paper studies an improved algorithm based on information filter, called the sparse extended information filter (SEIF). The SEIF algorithm is deduced by the sparsification treatment to EIF algorithm, which is the information form of EKF. This paper derives the structure of the SEIF algorithm in detail, and then analyzes the algorithm model. The algorithm analysis shows that the computational complexity of the SEIF algorithm is a constant, which is independent of environment features. That means SEIF has a high value of application in large-scale environment with a large number of features.After the model of SEIF based SLAM algorithm is set up, the simulation and experimental analysis of the algorithm is then given in this paper. The simulation results show that the estimation is accurate enough and do not induce large estimation error from sparsification treatment, which indicates the validity of the algorithm. The validity of the algorithm is further certified by the experimental research using the Car Park real environment experimental dataset. SEIF is compared with EKF by simulation and experiment in this paper, and the results show that although the estimation error of SEIF is a little larger than EKF, SEIF is far better than EKF in terms of computation time and memory usage. That shows the superiority of SEIF compared to EKF in the computational complexity.