Learn More
A multilayered neural network processor is used to control a given plant. Several learning architectures are proposed for training the neural controller to provide the appropriate inputs to the plant so that a desired response is obtained. A modified error back propagation algorithm, based on propagation of the output error through the plant, is introduced.(More)
Motor adaptation to a novel dynamic environment is primarily thought of as a process in which the nervous system learns to anticipate the environmental forces to eliminate kinematic error. Here we show that motor adaptation can more generally be modeled as a process in which the motor system greedily minimizes a cost function that is the weighted sum of(More)
— We develop a numerically efficient algorithm for computing controls for nonlinear systems that minimize a quadratic performance measure. We formulate the optimal control problem in discrete-time, but many continuous-time problems can be also solved after discretization. Our approach is similar to sequential quadratic programming for finite-dimensional(More)
A new way to look at the learning algorithm in the cerebellar model articulation controller (CMAC) proposed by J.S. Albus (1975) is presented. A proof that the CMAC learning always converges with arbitrary accuracy on any set of training data is obtained. An alternative way to implement CMAC based on the insights obtained in the process is proposed. The(More)
— In this paper a state estimation technique is developed for sensing inclination angles using relatively low cost sensors. A low bandwidth tilt sensor is used along with an inaccurate rate gyro to obtain the measurement. The rate gyro has an inherent bias along with sensor noise. The tilt sensor uses an internal pendulum and therefore has its own slow(More)
  • S Chang, X L Chao, M Pinedo, J G Shanthikumar, Stochastic, Y C Chow +6 others
  • 2004
1497 IV. CONCLUSIONS In this paper we have considered optimal routing policies that make use of limited state information, in queueing systems with finite capacities. There are two motivations for studying such policies. First, it is not always realistic to assume that the queue lengths are observed. In this case, the system's manager has to specify a(More)
Abstmct-We investigate approaches to balance control of an under-actuated robot. The robot is similar in structure t o the Acrobot, but it is actuated with a pneumatic cylinder rather than an electric motor, and it is capable of hopping. Regions of attraction of the control system are studied, and two methods are presented that increase the size of this(More)
This paper addresses issues concerning the development of Virtual Reality representations of archaeological data from the viewpoint of the archaeologist and the educator, as these are seen through the authors' experience of constructing immersive virtual archaeology programmes for the broad public. Specifically, the issue of interactivity is approached(More)