A parallel algorithm, architecture and FPGA realization for landmark determination and map construction in a planar unknown environment

Good representations of the environment are crucial for navigation of autonomous systems. This paper presents a time and space-efficient hardware-based solution for construction of an environment map based on range sensor data gathered by a robot at P locations in an unknown 2D environment. In particular, the paper proposes a new parallel algorithm based on training a set of K place units for generating ''perceptual landmarks'' and connecting the landmarks based on sensor information. The computational complexity of this parallel algorithm is O(MPlogK) where M is the number of training iterations. The algorithm has been mapped to hardware. The architecture comprises of components such as Content Addressable Memory and Wallace tree adders. The architecture has been implemented in Xilinx Field Programmable Gate Array (FPGA) and results show that the proposed design achieves high speed taking approximately 100ms to identify 49 landmarks and build a map of an environment of size 3.75mx3.75m. Further, the entire design for an environment with 196 sample points and 49 processing elements fits in one XCV3200E (or XC2V6000) device. The details of the experimental setup for gathering data are also presented.

[1]  T. Ogura,et al.  A 20 kbit associative memory LSI for artificial intelligence machines , 1989 .

[2]  Daniel S. Katz,et al.  NASA Advances Robotic Space Exploration , 2003, Computer.

[3]  André DeHon,et al.  The Density Advantage of Configurable Computing , 2000, Computer.

[4]  Ulrich Nehmzow,et al.  Landmark-based navigation for a mobile robot , 1998 .

[5]  Jan M. Rabaey,et al.  Digital Integrated Circuits , 2003 .

[6]  S. R. Smith,et al.  An FPGA-based custom coprocessor for automatic image segmentation applications , 1994, Proceedings of IEEE Workshop on FPGA's for Custom Computing Machines.

[7]  Randall D. Beer,et al.  Spatial learning for navigation in dynamic environments , 1996, IEEE Trans. Syst. Man Cybern. Part B.

[8]  Teuvo Kohonen,et al.  The self-organizing map , 1990 .

[9]  Sergio D'Angelo,et al.  A fault-tolerant FPGA-based multi-stage interconnection network for space applications , 2002, Proceedings First IEEE International Workshop on Electronic Design, Test and Applications '2002.

[10]  Thambipillai Srikanthan,et al.  High-Speed Environment Representation Scheme for Dynamic Path Planning , 2001, J. Intell. Robotic Syst..

[11]  Thambipillai Srikanthan,et al.  A linear approximation based hybrid approach for binary logarithmic conversion , 2002, Microprocess. Microsystems.

[12]  Anwar S. Dawood,et al.  FPGA based real-time adaptive filtering for space applications , 2002, 2002 IEEE International Conference on Field-Programmable Technology, 2002. (FPT). Proceedings..

[13]  T. Ogura,et al.  A 4-Kbit associative memory LSI , 1985, IEEE Journal of Solid-State Circuits.

[14]  Lawrence Chisvin,et al.  Content-addressable and associative memory: alternatives to the ubiquitous RAM , 1989, Computer.

[15]  Russell Greiner,et al.  Learning to Select Useful Landmarks , 1994, AAAI.

[16]  Leslie Pack Kaelbling,et al.  Uncertainty in Graph-Based Map Learning , 1993 .

[17]  Ken Hughes,et al.  A parallel algorithm and architecture for robot path planning , 1994, Proceedings of 8th International Parallel Processing Symposium.

[18]  John P. Hayes,et al.  Computer Architecture and Organization , 1980 .