Adaptation and Learning in Robotics and Automation
The paper attempts to describe the reciprocal impact that adaptive control, learning theory,.
Robotics and Automation↗
Engineering topics
Venkataraman, S. T.
Publications and source records attributed to Venkataraman, S. T..
Surface transport vehicles and supporting technology requirements
Requirements have been identified for surface transport vehicles which allow remote scientific exploration on the moon, as well as lunar resource recovery and emplacement of a permanent base on the lunar surface. Attention is given to the results of a design study which developed configurational concepts for lunar surface transport vehicles and inferred technology-development requirements, with a view to a phased program of implementation. Distinct benefits are noted for the design of simple vehicle platforms with high commonality, in order to reduce logistical-support requirements and maximize functional flexibility. Two generic vehicle classed are defined.
Matijevic, J. R.↗
Control of nonlinear systems using terminal sliding modes
The development of an approach to control synthesis for robust robot operations in unstructured environments is discussed. To enhance control performance with full model information, the authors introduce the notion of terminal convergence and develop control laws based on a class of sliding modes, denoted as terminal sliders. They demonstrate that terminal sliders provide robustness to parametric uncertainty without having to resort to high-frequency control switching, as in the case of conventional sliders. It is shown that the proposed method leads to greater guaranteed precision in all control cases discussed.
Venkataraman, S. T.↗
Compliance control with embedded neural elements
The authors discuss a control approach that embeds the neural elements within a model-based compliant control architecture for robotic tasks that involve contact with unstructured environments. Compliance control experiments have been performed on actual robotics hardware to demonstrate the performance of contact control schemes with neural elements. System parameters were identified under the assumption that environment dynamics have a fixed nonlinear structure. A robotics research arm, placed in contact with a single degree-of-freedom electromechanical environment dynamics emulator, was commanded to move through a desired trajectory. The command was implemented by using a compliant control strategy.
Venkataraman, S. T.↗
A representation for error detection and recovery in robot task plans
A general definition is given of the problem of error detection and recovery in robot assembly systems, and a general representation is developed for dealing with the problem. This invariant representation involves a monitoring process which is concurrent, with one monitor per task plan. A plan hierarchy is discussed, showing how diagnosis and recovery can be handled using the representation.
Lyons, D. M.↗
Design and implementation of a robot control system with traded and shared control capability
Preliminary results are reported from efforts to design and develop a robotic system that will accept and execute commands from either a six-axis teleoperator device or an autonomous planner, or combine the two. Such a system should have both traded as well as shared control capability. A sharing strategy is presented whereby the overall system, while retaining positive features of teleoperated and autonomous operation, loses its individual negative features. A two-tiered shared control architecture is considered here, consisting of a task level and a servo level. Also presented is a computer architecture for the implementation of this system, including a description of the hardware and software.
Hayati, S.↗