Online Feature Weighting for Human Discrimination in a Person Following Robot - Foundations on Natural and Artificial Computation - page 230

Fig. 6. Evolution of feature weights during the second sequence

each feature included in the feature space and enhances the weight according to
the score obtained by each feature. This increases the dissimilarity between the
target being followed by the robot, and the distractors moving close to it. The
experimental results show the high performance of our proposal working on a
real robot. The person-following behaviour was also tested during large periods
of time on crowded environments showing the expected behaviour.

Acknowledgements

This work was supported by the research grants TIN2009-07737 and

INCITE08PXIB262202PR.

References

1. Gockley, R., Forlizzi, J., Simmons, R.: Natural person-following behaviour for social
robots. In: Proceedings of the ACM/IEEE International Conference on Human-
robot Interaction, pp. 17-24 (2007)
2. Yoshimi, T., Nishiyama, M., Sonoura, T., Nakamoto, H., Tokura, S., Sato, H.,
Ozaki, F., Matsuhira, N., Mizoguchi, H.: Development of a person following robot
with vision based target detection. In:IEEE/RSJ International Conference on In-
telligent Robots and Systems, pp. 5286-5291 (2007)
3. Bellotto, N., Hu, H.: People tracking and identification with a mo-
bile robot. In: International Conference on Mechatronics and Automation,
ICMA 2007, pp. 3565-3570 (2007)
4. Bellotto, N., Hu, H.: Multisensor-based human detection and tracking for mobile
service robots. IEEE Transactions on Systems, Man, and Cybernetics, Part B:
Cybernetics 39(1), 167-181 (2009)
5. Dalal, N., Triggs, B.: Histograms of oriented gradients for human detection. In:
IEEE Computer Society Conference on Computer Vision and Pattern Recognition
CVPR 2005, vol. 1, pp. 886-893 (2005)
6. Ojala, T., Pietikainen, M., Harwood, D.: A comparative study of texture mea-
sures with classification based on featured distributions. Pattern Recognition 29(1),
51-59 (1996)