Robot Localization Using Particle Filter PDF

Chapter 9: Robot Localization Particle Filter  Mobilc Robot Localization problcm Robot localization is the process of determining where a mobile robot is located with respect to its environment. Localization is one of the most fundamental competencies required by an auto...

Chapter 9: Robot Localization Particle Filter  Mobilc Robot Localization problcm Robot localization is the process of determining where a mobile robot is located with respect to its environment. Localization is one of the most fundamental competencies required by an autonomous robot as the knowledge of the robot's own location is an essential precursor to making decisions about future actions. In a typical robot localization scenario, a map of the environment is available and the robot is equipped with sensors that observe the environment as well as monitor its own motion. The localization problem then becomes one of estimating the robot position and orientation within the map using information gathered from these sensors. Robot localization techniques need to be able to deal with noisy observations and generate not only an estimate of the robot location but also a measure of the uncertainty of the location estimate Sumplc buscd Localization (Sonur)  Mobile Robot Localization problcm Red points are hypotheses of possible state of the system, potential position of the robot. Noting that orientation is not shown, so red dots are x and y only Robot is located in point of the beginning of the blue line and directed by joystick to the room In the right top, a first scan before starting computations No doors are shown, this makes it harder to robot to find out where it is and which direction it is facing However, moving the robot to some room shall decrease the disambiguation the robot is facing, moreover, scans may help the robot to figure up the direction of the door as some of the sonar beams are indicating obstacles while others not.  Particle filters Particle filters are a way to efficiently represent non Gaussian distribution Basic principle Set of state hypotheses particles Survival of the fittest potential state position of a system robot Particle Filters Represent belief by random samples Estimation of non Gaussian, nonlinear processes Monte Carlo filter, Survival of the fittest, Condensation, Bootstrap filter, Particle filter Mathematical Description Set of weighted samples The samples represent the posterior Function Approximution Particle sets can be used to approximate functions The more particles fall into an interval, the higher the probability of that interval How to draw samples form a function/distribution? Rejection sampling Rejection sampling is a method to generate samples from a pdf that is unnormalized Importance Sampling Principle Importance sampling with Resampling We pass particles to motion model and then weight them according to observation model Particle Filter Algorithm Sample the next generation for particles using the proposal distribution. Compute the importance weights Weight = target distribution / proposal distribution Resampling: “ Replace unlikely samples by more likely ones” Mobile Robot Localization Each particle is a potential pose of the robot Proposal distribution is the motion model of the robot (prediction step) The observation model is used to compute the importance weight  (correction step) The set of weighted particles approximates the posterior belief about the robot's pose (target distribution) Particle Filter Algorithm Particle Filter Algorithm Resampling Mobilc Robot Localization Each particle is a potential pose of the robot Proposal distribution is the motion model of the robot (prediction step) The observation model is used to compute the importance weight  (correction step) The set of weighted particles approximates the posterior belief about the robot's pose (target distribution) Mobilc Robot Localization Using Particle Filter Mobilc Robot Localization Using Particle Filter Localization Example