TP 19

The 8th 2011 8th International International Conference Conference on on Ubiquitous Ubiquitous Robots Robots and and Ambient Ambient Intelligence Intelligence (URAI (URAI)2011) Nov. 23-26, 2011 in Songdo ConventiA, Incheon, Korea

‫ٻ‬

Topological Map Building for Mobile Robots Based on GIS in Urban Environments Yu-Cheol Lee1, Christiand1 , Seung-Hwan Park1, Wonpil Yu1, and Sung-Hoon Kim1 1

Department of Robot Cognition System, ETRI, Daejeon, 305-700, Korea (E-mail: yclee, christiand, sinkyv, ywp, [email protected])

Abstract - This paper presents an efficient method for building a topological map for robots in urban environments based on a geographic information system (GIS) such as satellite maps. In urban space, mobile robots need a special map, such as a topological map, to generate a path toward their goal. Unlike a car map, a topological map for mobile robot navigation should include semantic data, e.g., the width and type of road. This paper divides the GIS-based topological map building process into two steps. The first step is defining the topological map model to fit the mobile robot navigation in an urban environment. The second step is generating a topological map from existing GIS data to reduce the cost and improve the accuracy in building the map. Keywords - Topological map, GIS, navigation, mobile robot

1. Introduction Mobile robots need proper information regarding their environment, called a ‘robot map’, to accomplish their assigned tasks. A robot map is used to localize the positions of the robots, targets, and obstacles that might interrupt their movements. The positions of objects are stored in the form of a map by using sensor data and a mapping algorithm [1][2]. The outdoor navigation of mobile robots has some special issues. First, the robot navigates in a large-scale environment due to the nature of the outdoors. Second, in almost all cases except a few special situations, robots navigate only in specified regions such as roads. Based on the considerations mentioned above, the topological map is the most suitable map type for the outdoor navigation of mobile robots. In a topological map, a space is represented as linked nodes. Thus, the spatial information is acquired by interpreting the connectivity information between nodes. The region of interest is presented as a node, and the link is considered as the road and is used for generating the path. To build the topological map for mobile robots, we should define the form of the topological map so that it is applicable to any kind of mobile robot with various types of sensors. The robot map should be widely accepted for use in any kind of mobile robot. In this paper, we used a geographic information system (GIS) based on satellite images [3] to make a topological map. Using the satellite images, it is easy to gather geometric data and road information for the targeted outdoor environment. This paper defines the format of topological map and presents the method used to make a topological map with GIS and proposes some future works.

978-1-4577-0723-0 / 11 / $26.00 Ϳ 2011 IEEE

2. Format of Topological Map Topological map M is defined as two parts consisting of Origin and Segments in (1). M

^Origin, Segments`

(1)

Origin is the basic information required to build a topological map based on GIS, such as Mapsize, Offset, and GPS geo-location at the origin in (2). Origin

^Mapsize, Offset , GPS`

(2)

Mapsize includes information regarding the width, height, and pixel_size of the satellite image extracted from GIS in (3). pixel_size indicates the scale factor from the pixel domain to the metric domain. Mapsize

^width, height , pixel _ size`

(3)

An Offset is defined as the translation and rotation offsets of the origin point with respect to the left-bottom area of a satellite image. Offset consists of x, y, and  offsets in (4). GPS sensor, which is the most used sensor for localization systems in outdoor environments, provides robot position data based on the geodetic domain. Therefore, a topological map provides the GPS parameters in (5) for converting the position unit from geodetic to metric domains. In (6) and (7), GPS data is organized into hierarchies, which include Latitude, Longitude, and  measured by real GPS sensors at the origin point. Offset GPS

^ x, y , T `

^Latitude, Longitude, T `

Latitude = ^direction, degree, minute, second `lat

(4) (5) (6)

Longitude = ^direction, degree, minute, second `lon (7)

A topological map is designed as a hierarchical structure containing node information. A Segment is defined as a region in which a mobile robot may travel along a specific route. Segments include several nodes as shown in (8). Each Segment is comprised of one or more roads in (9). road is the subordinate of Segment that is used by path planner to generate a path. A road is characterized by the number of nodes in (10). A node is generally placed at the center of the pavement. Every node is specified by a unique identifier, position, GPS uncertainty, number of lanes, road width, speed limit of the mobile robot, type of traffic markers on the road, and connectivity among nodes.



Fig. 1. The steps used to assign map properties

^

Segments = S 1 , , S i

^

S i = R1i , , R ij

^

`

`

R ij = N ij ,1 , , N ij , k

(8) (9)

` i

N ij , k

Fig. 2. Node selection for a topological map

(10)

­id , x , y, T , uncertainty, lane _ num, ½ ® ¾ (11) ¯ width, vel , type, connect _ ids ¿ j ,k

robot to generate its path. In a typical urban environment, the nodes shape the road network (links) that show major common roads, as shown in Fig. 2. 3.4 Surveying

3. Topological Map Building 3.1 Preparation of Workspace Image A workspace image represents the surrounding environment as a single image where a robot is operated. The image is simply taken from an aerial view image from a satellite. These days, aerial view images can be widely accessed from the Internet, for example through GIS services such as Google Earth [3]. This method is without cost, and only needs an Internet connection. 3.2 Map Property Assignment After a workspace image is acquired, some map parameters should be assigned as in (2). This is related to an interpretation between image domain and metric domain (real world). The types of parameters to be assigned include image size, scale factor, and origin point, as shown in Fig. 1. Since we use a GPS as the main sensor, the origin point is represented as a geo-location (latitude and longitude). Image size is the parameter related to the size of the workspace image. The image size shows the resolution of the workspace image. Scale factor is a value that maps the location from the pixel domain to metric domain. This scale factor will be used throughout the operation whenever a robot needs to infer certain information from the topological map by giving its location, i.e., map matching for GPS measurement. 3.3 Nodes Selection Based on the map properties, the map maker can easily select the nodes for the topological map in the workspace. The nodes represent the topology relationships that inform the many possibilities for the

The purpose of surveying is to fill the information of each node, such as identification, location (x, y, and ), and level of uncertainty at the time of surveying. The level of uncertainty is useful information for the robot to make decisions for a task. For instance, if a particular node has a large uncertainty, it means that the robot should carefully use that information and not place great dependency on it. 3.5 Data Conditioning It is well known that data from surveying sometimes contain an outlier and false data. To avoid erroneous data, the map maker should filter such data out. Since we also include the level of uncertainty in the node information, we are able to improve the localization performance at the nodes through a simple threshold filtering of erroneous data.

4. Future Works To validate the proposed mapping method, we will show details of its implementation in map building as well as in a navigation experiment in real out door environments such as a campus or city district.

References [1] C. Jang, S. I. Lee, et al., “OPRoS: A New Component-Based Robot Software Platform,” ETRI Journal, vol. 32, no. 5, pp. 646-656, 2010. [2] S. Thrunet al., “A Real-time Expectation Maximization Algorithm for Acquiring Multiplanar Maps of Indoor Environments with Mobile Robots,” IEEE Transactions on Robotics and Automation, vol. 20, no. 3, pp. 433-443, 2004. [3] Google Earth. Available: http://earth.google.com



Topological Map Building for Mobile Robots Based on ...

Component-Based Robot Software Platform,” ETRI. Journal, vol. 32, no. 5, pp. ... IEEE Transactions on Robotics and Automation, vol. 20, no. 3, pp. 433-443 ...

506KB Sizes 1 Downloads 198 Views

Recommend Documents

Long–Term Topological Localisation for Service Robots in Dynamic ...
and for solving the kidnapping [3] problem in mobile robotics. Typical approaches have made use of static representations of the word to solve the robot localisation problem. However, service robots have to operate in populated environments that are

Long–Term Topological Localisation for Service Robots in Dynamic ...
logical localisation of service robots in dynamic indoor envi- ronments. ... large-scale global localization [1], loop-closure detection [2], and for solving the ..... its status regularly using a social network interface, so its occasional failures

Multi-Stage Localization Given Topological Map for ...
Faculty of Computer and Information Science,. Ain Shams University in Cairo ... The range of 360 degrees is divided into 6 bins. – The average length of lines in ...

Robust Obstacle Segmentation based on Topological ...
persistence diagram that gives a compact visual representation of segmentation ... the 3D point cloud estimated from the dense disparity maps computed ..... [25] A. Zomorodian and G. Carlsson, “Computing persistent homology,” in Symp. on ...

Appearance-Based Topological Bayesian Inference for ...
Based on this approach, the probabilistic Bayesian in- ference can work ...... loop-closing for SLAM in a highly unstructured cross-country environment where no ...

Fall Prediction of Legged Robots Based on Energy ...
Abstract—In this paper, we propose an Energy based Fall. Prediction (EFP) which observes the real-time balance status of a humanoid robot during standing.

On Reliable Topological Structures for Message ...
Paper approved by the Editor for Computer Communication of the ... C-0609 and in part by the National Science Foundation under Grant. The author is with the ..... by (noting that in H(n - 1, k - 1) every vertex has degree ... (c) The Graph BA.

Motion planning for formations of mobile robots
if, for example, distributed data needs to be collected at different resolutions and thus ... advantageous, for example, in a data collection task where a particular ...

Motion planning for formations of mobile robots - Semantic Scholar
for tele-operation of such a network from earth where communication .... A clear advantage of using motion planning for dynamic formations is that maneuvers can .... individual robots via a wireless RC signal. ... may provide a disadvantage for appli

Stair Climbing Mechanism For Mobile Robots (Msrox)
MSRox with 83 cm in Length, 54 cm in Width and 29 cm in Height has been designed to climb stairs with 10 cm in .... robot can go back to the state of flat motion.

Autonomous Mobile Robots-siegwart.pdf
The idea of providing a robot. functional architecture as an outline of the book, and then explaining each. component in a chapter, is excellent. I think the authors ...

Low-Power Log-MAP Decoding Based on Reduced ...
decide the termination point [10], [11]. Though the early termi- nation is one .... pute LLR values. In this paper, we call this case an approxima- tion failure. 2) and.

airberlin introduces a new route map based on Google Maps
Commerce Business Development Manager at airberlin. Möscher notes that with the ... a short introduction to the Google Maps interface and Analytics for Maps. API for Work. ... Use of advanced geocoding functions with a larger volume and ...

Simultaneous Topological Map Prediction and Moving ...
Abstract— To achieve fully autonomous mobile robot in unknown environment, a mapping and localization ... Moving Object Tracking, Mobile Robot. I. INTRODUCTION. CENE understanding is one of the most ..... are affected by the velocity and steering o

Port-based Modeling and Control for Efficient Bipedal Walking Robots
(e.g. series or parallel connection for the damper), without the need to rewrite the ... control in terms of interacting physical systems has several advantages. ...... board and an external PC by wireless Bluetooth communication or a wired serial.

Port-based Modeling and Control for Efficient Bipedal Walking Robots
3.3.4 Conditions for contact release . ..... robot, and Wisse & van Frankenhuyzen (2003) extended McGeer's original pas- .... 1.2. PORT-HAMILTONIAN MODELING AND CONTROL. 7. 0 m. 30 m. 60 m ...... Photoshop and even LaTeX! Besides ...

Topological Synthesis of Mobile Backbone Networks ... - Springer Link
topological synthesis and management of Mobile Backbone Network (MBN) based architectures. ... and video, notwithstanding the lack of a fixed infrastructure.

Topological Synthesis of Mobile Backbone Networks ... - Springer Link
topological synthesis and management of Mobile Backbone Network (MBN) .... accessibility (i.e., a certain percentage of nodes should be within a fixed small.

FreMEn: Frequency Map Enhancement for Long-Term Mobile Robot ...
port for long-term mobile robot autonomy. Recent works have demonstrated that exploiting the outlying measurements allows to characterize some environment changes, which improves robot localisation in changing environments [3], [4], [5], [6]. In our

evaluation of future mobile services based on the ... - Semantic Scholar
However, most of the mobile data services have fallen short of the expectation and have ..... Journal of the Academy of Marketing Science,. 33 (3), pp. 330-346.

evaluation of future mobile services based on the technology ...
Keywords: Mobile Broadcast Services, Technology Acceptance Model ... The proposed TAM model was widely accepted by the information system research ..... Antecedents and Cross-Service Comparisons, Journal of the Academy of ...

FreMEn: Frequency Map Enhancement for Long-Term Mobile Robot ...
spectral model to the time domain allows for the prediction of the future environment .... free, edges of a topological map are traversable or not, doors are open or ...

Disseminating Active Map Information to Mobile Hosts
Each of the approaches exhibits di erent trade-o s of server, network, and client loads. To ..... subscribers via a single multicast channel that is dedicated to them.