Abstract：A novel method for material identification based on electrostatic signal detection technology is presented. 4 kinds of typical roads, i.e. brick, sand, grass and asphalt, which can be often encountered in outdoor environment, are effectively identified using the proposed method. The induced charge change on robot foot is analyzed by establishing an equivalent model for the contact/separation process between the robot foot and the road surface. The simulation result shows that there are obvious differences in the discharge of surface charge of different pavement materials. Based on that, a special structure of measurement platform is proposed for simulation of contact and separation between robot foot and roads. 4 kinds of road surface electrostatic signals are collected, and the feature value of the signal is extracted as the classifier parameter. The k-nearest neighbor classifier is used to classify the road surface electrostatic signals. The result shows that the overall recognition rate is about 83.3%.
[1] Filitchkin P, Byl K. Feature-based terrain classification for LittleDog[C]//IEEE/RSJ International Conference on Intelligent Robots and Systems. Piscataway, USA:IEEE, 2012:1387-1392.
[2] Khan Y N, Komma P, Bohlmann K, et al. Grid-based visual terrain classification for outdoor robots using local features[C]//IEEE Symposium on Computational Intelligence in Vehicles and Transportation Systems. Piscataway, USA:IEEE, 2011:16-22.
[3] 李强,薛开,徐贺,等.基于振动采用支持向量机方法的移动机器人地形分类[J].机器人,2012,34(6):660-667.Li Q, Xue K, Xu H, et al. Vibration-based terrain classification for mobile robots using support vector machine[J]. Robot, 2012, 34(6):660-667.
[4] Kim K, Ko K, Kim W, et al. Performance comparison between neural network and SVM for terrain classification of legged robot[C]//49th Annual Conference of the Society of Instrument and Control Engineers of Japan. Piscataway, USA:IEEE, 2010:1343-1348.
[5] Hoepflinger M A, Remy C D, Hutter M, et al. Haptic terrain classification for legged robots[C]//IEEE International Conference on Robotics and Automation. Piscataway, USA:IEEE, 2010:2828-2833.
[6] Ojeda L, Borenstein J, Witus G, et al. Terrain characterization and classification with a mobile robot[J]. Journal of Field Robotics, 2006, 23(2):103-122.
[7] Lowell J, Rose-Innes A C. Contact electrification[J]. Advances in Physics, 1980, 29(6):947-1023.
[8] Lowell J, Akande A R. Contact electrification——Why is it variable?[J]. Journal of Physics D:Applied Physics, 1988, 21(1):125-137.
[9] Tada Y, Inoue M, Kawasaki T, et al. A flexible and stretchable tactile sensor utilizing static electricity[C]//IEEE/RSJ International Conference on Intelligent Robots and Systems. Piscataway, USA:IEEE, 2007:684-689.
[10] Tada Y, Inoue M, Kawasaki T, et al. A principle and characteristics of a flexible and stretchable tactile sensor based on static electricity phenomenon[J]. Journal of the Robotics Society of Japan, 2008, 26(2):210-216.
[11] Kimoto A, Ichinose Y, Shida K. A new sensing method using contact voltage for material discrimination[J]. Sensors and Actuators A:Physical, 2009, 149(1):1-6.
[12] Kimoto A, Sugitani N. A new sensing method based on PVDF film for material identification[J]. Measurement Science and Technology, 2010, 21(7):No.075202.
[13] Kurita K. New approach to estimate friction caused by biped robot walking based on electrostatic induction[C]//International Conference on Advanced Mechatronic Systems. Piscataway, USA:IEEE, 2012:680-683.
[14] Raibert M, Blankespoor K, Nelson G, et al. BigDog, the rough-terrain quadruped robot[J]. IFAC Proceedings Volumes, 2008, 41(2):10822-10825.
[15] Chen X, Zheng Z, Cui Z Z, et al. A novel remote sensing technique for recognizing human gait based on the measurement of induced electrostatic current[J]. Journal of Electrostatics, 2012, 70(1):105-110.
[16] Xue K, Li Q, Xu H, et al. Vibration-based terrain classification for robots using k-nearest neighbors algorithm[J]. Journal of Vibration, Measurement and Diagnosis, 2013, 33(1):88-92.
