ON TERRAIN MODEL ACQUISITION BY A POINT ROBOT AMIDST POLYHEDRAL OBSTACLES

被引：18

作者：

RAO, NSV

IYENGAR, SS

OOMMEN, BJ

KASHYAP, RL

机构：

[1] LOUISIANA STATE UNIV,DEPT COMP SCI,BATON ROUGE,LA 70803

[2] CARLETON UNIV,SCH COMP SCI,OTTAWA K1S 5B6,ONTARIO,CANADA

[3] PURDUE UNIV,DEPT ELECT ENGN,W LAFAYETTE,IN 47907

来源：

IEEE JOURNAL OF ROBOTICS AND AUTOMATION | 1988年 / 4卷 / 04期

关键词：

Another interesting problem ia the navigation of a robot in an unexplored or a partially explored terrain. In this case; the entire terrain model may not be known; and the robot relies on its sensors for navigation. Lumelsky and Stepanov 141 present sensor-based navigation algorithms for navigating a point automaton to a destination point using touch type of sensor. In this inethod localized sensor information is used to guide the point automaton; and this information is not put to any further global use. In many applications; incidental learning is shown to be an important enhancement in the navigation planning. Here; a composite model of the terrain is built by integrating the sensor information obtained as the robot executes sensor-based and goal-directed navigation. Iyengar et al. [2; Oommen et al. [6; Turchan and Wong 191 discuss different versions of learned navigation in unexplored terrains. Here we consider the problem of acquiring the terrain model by systematic exploration of the terrain using a sensor. Our main motivation stems from the fact that the availability of the terrain model enables us to plan the entire Manuscript received February 26; 1987; revised December 7. 1987. A preleminary version of this paper was presented at the 3rd IEEE Conference on AI Applications; Orlando; FL. Feb. 1987. The work of B. J. Oomman was partially supported by the National Sciences and Engineering Council of Canada. N. S. V. Rao is with the Department of Computer Science; Old Dominion University; Norfolk; VA 23529-0162. S. S. Iyengar is with the Department of Com:)uter Science; Louisiana State University; Baton Rouge; L.4 70803. B. J. Oommen is with the School of Computer Science. Carleton University; Ottawa KIS SB6; Canada. R. J. Kashyap is with the Department of Electrical Engineering. Purdue University; West Lafayette; IN 47907. IEEE Log Number 8820163;

D O I：

10.1109/56.812

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

引用

页码：450 / 455

页数：6

共 11 条

[1] SHAPE FROM PROBING [J].