Search

This paper presents a method for SLAM in Robot and Wireless Sensor Network (WSN) System using Bayes framework. A mobile robot equipped with sensor measuring the range to WSN nodes by Received Signal Strength Indicator (RSSI) can simultaneously localize itself as well as locate the sensor nodes of WSN. We adopted an algorithm based Rao-Blackwellized Particle Filter (RBPF) to integrate measurements from the different nodes over time while the robot movies in the environment. Each particle contains an estimation robot pose and a set of auxiliary filters estimating the position of sensor nodes, one for each sensor node which had been observed by robot. The auxiliary filter can switch from Particle Filter in the initial stage to EKF in the subsequent stage due to the characteristic of range-only measurement. The experiment proved the efficiency and practicality of the algorithm.

A new laser scan matching method based on clustering is advanced. It takes the advantages of both non-iterative matching method and iterative matching method as well as increases the speed and the precision of matching. A method of map building based on laser scans using the matching method above is presented for mobile robots. The map building method matches two consecutive scans by using the feature points extracted from the scans and then completes the whole map building. In the experiment using the rescue robot developed in our lab, it shows that the new method can accomplish the task of real-time map building in an indoor environment accurately and effectively.

One of the challenges of SLAM (Simultaneous Localization and Mapping) for autonomous robots is the loop closing problem. In this paper, a decision-theoretic active loop closing approach is presented, which integrates the exploration planning with loop closing. In our approach, the active loop closing process is modeled as a multi-stage decision problem, and a frontier-based auxiliary topological map is build to assist the decision process. The autonomous robot chooses its actions according to the sequential decision results. The unknown range most likely to close a loop is selected to explored, and a particle-filter-based localization and smoothing method applied to partial maps is used in the loop validating and loop constraints building process. Experiments have shown that our approach can practically implement loop closure and obviously improve the mapping precision compared to passive exploration strategy.

This paper illustrates the reason why revisable data association is needed for the simultaneous localization and mapping (SLAM) of mobile robots, and an incremental SLAM algorithm with backtrack searching data association is presented. Our approach uses a tree model called correspondence tree (CT) to represent the solution space of the data association problem. CT is layered according to time steps and every node in it is a data association hypothesis for the measurements gotten at-a-time. A best-first with limit backtracking search strategy is designed to find the optimal path in CT. A state estimation method based on the least-squares problem is developed. This method can compute the cost of nodes in CT and update state estimation incrementally, so direct feedback is introduced from the state estimation process to the data association model. With the interaction between data association and state estimation, and combining with tree pruning techniques, our approach can get accurate data association and state estimation for online SLAM applications. The contribution of this paper is that we have analyzed the necessity of revisable data association for SLAM and we use graph search of AI to model and solve the revising data association problem.

This paper focuses on the reciprocity between data association and state estimation in the simultaneous localization and mapping (SLAM) problem for mobile robot, and an incremental SLAM algorithm with backtracking data association is presented. Our approach uses a tree model called correspondence tree (CT) to represent the solution space of the data association problem. CT is layered according to time steps and every node in it is a data association hypothesis for the measurements gotten at a time. A best-first with limit backtracking search strategy is designed to find the optimal path in CT. A state estimation method based on the least-squares problem is developed. This method can compute the cost of nodes in CT and update state estimation incrementally, so direct feedback is introduced from the state estimation process to the data association model. With the interaction between data association and state estimation, and combining with tree pruning techniques, our approach can get accurate data association and state estimation for on-line SLAM applications.