Roles of Knowledge in Motor Learning

The goal of this thesis is to apply the computational approach to motor learning, i.e., describe the constraints that enable performance improvement with experience and also the constraints that must be satisfied by a motor learning system, describe what is being computed in order to achieve learning, and why it is being computed. The particular tasks used to assess motor learning are loaded and unloaded free arm movement, and the thesis includes work on rigid body load estimation, arm model estimation, optimal filtering for model parameter estimation, and trajectory learning from practice. Learning algorithms have been developed and implemented in the context of robot arm control. The thesis demonstrates some of the roles of knowledge in learning. Powerful generalizations can be made on the basis of knowledge of system structure, as is demonstrated in the load and arm model estimation algorithms. Improving the performance of parameter estimation algorithms used in learning involves knowledge of the measurement noise characteristics, as is shown in the derivation of optimal filters. Using trajectory errors to correct commands requires knowledge of how command errors are transformed into performance errors, i.e., an accurate model of the dynamics of the controlled system, as is demonstrated in the trajectory learning work. The performance demonstrated by the algorithms developed in this thesis should be compared with algorithms that use less knowledge, such as table based schemes to learn arm dynamics, previous single trajectory learning algorithms, and much of traditional adaptive control.

Motor Control and Learning by the State Space Model

A model is presented that deals with problems of motor control, motor learning, and sensorimotor integration. The equations of motion for a limb are parameterized and used in conjunction with a quantized, multi-dimensional memory organized by state variables. Descriptions of desired trajectories are translated into motor commands which will replicate the specified motions. The initial specification of a movement is free of information regarding the mechanics of the effector system. Learning occurs without the use of error correction when practice data are collected and analyzed.

A Systems Approach to the Torque Control of a Permanent Magnet Brushless Motor

Many approaches to force control have assumed the ability to command torques accurately. Concurrently, much research has been devoted to developing accurate torque actuation schemes. Often, torque sensors have been utilized to close a feedback loop around output torque. In this paper, the torque control of a brushless motor is investigated through: the design, construction, and utilization of a joint torque sensor for feedback control; and the development and implementation of techniques for phase current based feedforeward torque control. It is concluded that simply closing a torque loop is no longer necessarily the best alternative since reasonably accurate current based torque control is achievable.

Lean Transition of Emerging Industrial Capability

This volume of the final report documents the technical work performed from December 1998 through December 2002 under Cooperative Agreement F33615-97-2-5153 executed between the U.S. Air Force, Air Force Research Laboratory, Materials and Manufacturing Directorate, Manufacturing Technology Division (AFRL/MLM) and the McDonnell Douglas Corporation, a wholly-owned subsidiary of The Boeing Company. The work was accomplished by The Boeing Company, Phantom Works, Huntington Beach, St. Louis, and Seattle; Ford Motor Company; Integral Inc.; Sloan School of Management in the Massachusetts Institute of Technology; Pratt & Whitney; and Central State University in Xenia, Ohio and in association with Raytheon Corporation. The LeanTEC program manager for AFRL is John Crabill of AFRL / MLMP and The Boeing Company program manager is Ed Shroyer of Boeing Phantom Works in Huntington Beach, CA. Financial performance under this contract is documented in the Financial Volume of the final report.

Effects of Air Pollution Control on Climate

Urban air pollution and climate are closely connected due to shared generating processes (e.g., combustion) for emissions of the driving gases and aerosols. They are also connected because the atmospheric lifecycles of common air pollutants such as CO, NOx and VOCs, and of the climatically important methane gas (CH4) and sulfate aerosols, both involve the fast photochemistry of the hydroxyl free radical (OH). Thus policies designed to address air pollution may impact climate and vice versa. We present calculations using a model coupling economics, atmospheric chemistry, climate and ecosystems to illustrate some effects of air pollution policy alone on global warming. We consider caps on emissions of NOx, CO, volatile organic carbon, and SOx both individually and combined in two ways. These caps can lower ozone causing less warming, lower sulfate aerosols yielding more warming, lower OH and thus increase CH4 giving more warming, and finally, allow more carbon uptake by ecosystems leading to less warming. Overall, these effects significantly offset each other suggesting that air pollution policy has a relatively small net effect on the global mean surface temperature and sea level rise. However, our study does not account for the effects of air pollution policies on overall demand for fossil fuels and on the choice of fuels (coal, oil, gas), nor have we considered the effects of caps on black carbon or organic carbon aerosols on climate. These effects, if included, could lead to more substantial impacts of capping pollutant emissions on global temperature and sea level than concluded here. Caps on aerosols in general could also yield impacts on other important aspects of climate beyond those addressed here, such as the regional patterns of cloudiness and precipitation.