The fuzzy Kalman filter: State estimation using possibilistic techniques (original) (raw)
Related papers
A novel fuzzy Kalman filter for mobile robots localization
A new method to implement fuzzy Kalman filters is introduced in this paper. This has special application in fields where inaccurate models or sensors are involved, such as in mobile robotics. The innovation consists in using possibility distributions, instead of gaussian distributions. The main advantage of this approach is that uncertainty is not needed to be symmetric, while a region of possible solutions is allowed. The contribution of this work also includes a method to propagate uncertainty through both the process and the observation models. This one is based on quantifying uncertainty as trapezoidal possibility distributions. Finally, the way to reduce the EKF inconsistence when large number of iterations are carried out is shown.
Fuzzy logic based nonlinear Kalman filter applied to mobile robots modelling
2004 IEEE International Conference on Fuzzy Systems (IEEE Cat. No.04CH37542)
In order to reduce the false alarms in fault detection systems for mobile robots, accurate state estimation is needed. Through this work, a new method for localization of a mobile robot is presented. First, a Takagi-Sugeno fuzzy model of a mobile robot is determined, which is optimized using genetic algorithms, creating a precise representation of the kinematic equations of the robot. Then, the fuzzy model is used to design a new extension of the Kalman filter, based on several linear Kalman filters. Finally, the fuzzy filter is compared to the conventional extended Kalman filter, showing an improvement over the estimation made. The fuzzy filter also presents advantages in implementation, due to the fact that the covariance matrices needed are easier to estimate, increasing the estimation frequency. I.
The Possibilistic Kalman Filter: Definition and Comparison With the Available Methods
IEEE Transactions on Instrumentation and Measurement, 2021
The Kalman filter is a commonly used algorithm for predicting the state variables of a system. It is based on the model of the system and some measurements (observed over time), which are characterized by their own uncertainty. This paper defines a possibilistic Kalman filter, whose main feature is to predict the values of the state variables and the associated uncertainty, also when uncertainty contributions of non-random nature are present. This possibilistic Kalman filter is defined in the mathematical framework of the possibility theory and employes Random-Fuzzy variables and the related mathematics, since these variables can properly represent measurement results together with the associated uncertainty. A comparison with the available methods is provided, as well as a final validation.
Localization using fuzzy and Kalman filtering data fusion
In this paper a new localisation algorithm for the navigation of a mobile robot in outdoor environments is presented. The algorithm combines together a classical Kalman filter with a set of Fuzzy rules to fuse the information coming from different sensors. Some preliminary experimental results confirm the validity of the proposed approach.
Fuzzy Adaptive Particle Filter for Localization of a Mobile Robot
Lecture Notes in Computer Science, 2007
Localization is one of the important topics in robotics and it is essential to execute a mission. Most problems in the class of localization are due to uncertainties in the modeling and sensors. Therefore, various filters are developed to estimate the states in noisy information. Recently, particle filter is issued widely because it can be applied to a nonlinear model and a non-Gaussian noise. In this paper a fuzzy adaptive particle filter is proposed, whose basic idea is to generate samples at the high-likelihood using a fuzzy logic approach. The method brings out the improvement of an accuracy of estimation. In addition, this paper presents the localization method for a mobile robot with ultrasonic beacon systems. For comparison purposes, we test a conventional particle filter method and our proposed method. Experimental results show that the proposed method has better localization performance.
Hybrid Kalman Filter/Fuzzy Logic based Position Control of Autonomous Mobile Robot
International Journal of Advanced Robotic Systems, 2005
This paper describes position control of autonomous mobile robot using combination of Kalman filter and Fuzzy logic techniques. Both techniques have been used to fuse information from internal and external sensors to navigate a typical mobile robot in an unknown environment. An obstacle avoidance algorithm utilizing stereo vision technique has been implemented for obstacle detection. The odometry errors due to systematic-errors (such as unequal wheel diameter, the effect of the encoder resolution etc.) and/or non-systematic errors (ground plane, wheel-slip etc.) contribute to various motion control problems of the robot. During the robot moves, whether straight-line and/or arc, create the position and orientation errors which depend on systematic and/or non-systematic odometry errors. The main concern in most of the navigating systems is to achieve the real-time and robustness performances to precisely control the robot movements. The objective of this research is to improve the pos...
Fuzzy uncertainty modeling for grid based localization of mobile robots
International Journal of Approximate Reasoning, 2010
This paper presents a localization method using fuzzy logic to represent the different facets of uncertainty present in sensor data. Our method follows the typical predict-update cycle of recursive state estimators to estimate the robot's location. The method is implemented on a fuzzy position grid, and several simplifications are introduced to reduce computational complexity. The main advantages of this fuzzy logic method compared to most current ones are: (i) only an approximate sensor model is required, (ii) several facets of location uncertainty can be represented, and (iii) ambiguities in the sensor information are directly represented, thus avoiding having to solve the data association problem separately. Our method has been validated experimentally on two different platforms, a legged robot equipped with vision and a wheeled robot equipped with range sensors. The experiments show that our method can solve both the tracking and the global localization problem. They also show that this method can successfully cope with ambiguous observations, when several features may be associated to the same observation, and with robot kidnapping situations. Additional experiments are presented that compare our approach with a state-of-the-art probabilistic method.
Robot Localization and Kalman Filters On finding your position in a noisy world
The robot localization problem is a key problem in making truly autonomous robots. If a robot does not know where it is, it can be difficult to determine what to do next. In order to localize itself, a robot has access to relative and absolute measurements giving the robot feedback about its driving actions and the situation of the environment around the robot. Given this information, the robot has to determine its location as accurately as possible. What makes this difficult is the existence of uncertainty in both the driving and the sensing of the robot. The uncertain information needs to be combined in an optimal way.
On the Kalman Filter Approach for Localization of Mobile Robots
2016
In this work we analyze robot motion given from the UTIAS Multi-Robot Dataset. The dataset contains recordings of robots wandering in a confined environment with randomly spaced static landmarks. After some preprocessing of the data, an algorithm based on the Extended Kalman Filter is developed to determine the positions of robots at every instant of time using the positions of the landmarks. The algorithm takes into account the asynchronous time steps and the sparse measurement data to develop its estimates. These estimates are then compared with the groundtruth data provided in the same dataset. Furthermore several methods of noise estimation are tested, which improve the error of the estimate for some robots.
Control Engineering Practice, 1999
Accurate estimates of mobile robot location, if available, can be used to improve the performance of a vehicle dynamics control system. To this purpose, the data provided by odometric and sonar sensors are here fused together by means of an extended Kalman "lter, providing robot position and orientation estimates at each sampling instant. To cope with the tracking of long trajectories, the performance of the "lter is improved by introducing an on-line fuzzy-rule-based adaptation scheme.