A Real-Time System for Small Animal Neurorobotics at Spinal or Cortical Levels.
S.F. Giszter, Corey B. Hart, Ubong Ime Udoekwere, Sergey N. Markin, C. Barbe
- 发表年份
- 2005
- 引用次数
- 7
摘要
We present a real-time system for small animal neurorobotics and neuroprosthetics at spinal or cortical levels. The system combines biomechanics, neural recording and a 3D robotics system including sensable devices phantoms, ATI 6 axis force transducers, cybernetics (bionic technology) cerebus neural recording system, computer boards buffered DAS16s, and an OPTOTRAK and 120 Hz camera system. The system allows real-time combination of neural data, force data and robot interaction at rates of 1 kilohertz, and allows full data recording. Up to 2 robots and 5 force sensors with 128 channels of neural data and 16 channels of EMG form the present core system. The force sensors and robot are controlled from a single machine using a dedicated Venturcom/Phar Lap ETS real time operating system. Elastic, viscous, translational and barrier constraint fields can be combined or recruited by neural activity. The main control loop can also support PID control. Neural data from up to 256 neurons (2 per channel, 128 channels) identified in real-time with multiple threshold windows on the cybernetics are delivered for robot control. Other data collection is synchronized by the main host. The robot(s) connect to rats or frogs through bone implants or with a saddle-harness arrangement
关键词
相关论文
Statistical Learning Theory
Yuhai Wu, Vladimir Vapnik
1999
Artificial intelligence: a modern approach
1995
Fractional Differential Equations
Igor Podlubný
2025
Applied Nonlinear Control
Jean-Jacques Slotine, Weiping Li
1991