PosPosterv4 - The Rossum Project...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
The Rossum Project – http://rossum.sourceforge.net Poster by Vassilis Varveropoulos Robot Localization and Map Construction Using Sonar Data Page 1 of 10 Robot Localization and Map Construction Using Sonar Data Vassilis Varveropoulos vassilis@users.sourceforge.net The Rossum Project http://rossum.sourceforge.net Introduction An important challenge in small-scale robotics is finding a robot's position when only limited sensor information is available. There are many technologies available for robot localization, including GPS, active/passive beacons, odometry (dead reckoning), sonar, etc. In each approach, however, improvements in accuracy come at the cost of expensive hardware and additional processing power. For the robotics enthusiast, the key to successful localization is getting the best results out of cheap and widely available sensors. This paper presents a method for localization and map construction of a mobile robot using data from a sonar-based range sensor. No prior knowledge of the environment is assumed. The map is constructed autonomously by the robot. This method has been implemented and tested using both the Rossum Playhouse simulator (see reference [3]) and an enhanced Rug Warrior robot. Experimental results from both the simulator and the robot are presented at the end of this paper. This paper and the positioning application bundled with the simulator, is publicly available from [1]. The source code will also be released. The Positioning Algorithm In its localization phase, the algorithm determines the robot’s position by correlating a local map (generated by a range sensor sweep), with a global map. While the global map can be supplied in advance, this algorithm does not require prior knowledge of the robot’s environment. Instead, it uses sensor data to construct the global map dynamically. The algorithm estimates the robot’s location by comparing the global and local maps. To do so, it computes positions called feasible poses, where the expected view of the robot approximately matches the observed range sensor data. It then selects a best fit from the feasible poses. To evaluate feasible poses efficiently, the algorithm represents the global map as an occupancy grid (a matrix of cells, each having a value that indicates whether that cell is empty or occupied). Using its sensors, the robot determines range vectors (distance and bearing from detected objects or features) which are then compared against all occupied cells in the grid. If a range vector can be overlaid on the grid without interference by other occupied cells, it indicates a feasible pose. In addition to its occupancy value, each cell in the grid is also assigned a certainty value indicating the likelihood that the robot is located at that position. Each time a feasible pose is identified, the certainty value of the corresponding cell is incremented. After all feasible poses are considered, the grid cell with the highest certainty value is selected as the robot’s present
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 2
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 08/03/2010 for the course MECHANIC 65921 taught by Professor Jons during the Spring '10 term at Tampa.

Page1 / 10

PosPosterv4 - The Rossum Project...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online