Real-time control of systems with unknown and varying time-delays, using neural networks

AbstractThe remote operation of an asset with time-delays short enough to allow for ‘real-time’ or near real-time control – often referred to as low-latency teleoperations (LLT) – has important potential to address planetary protection concerns and to enhance astrobiology exploration. Not only can LLT assist with the search for extraterrestrial life and help mitigate planetary protection concerns as required by international treaty, but it can also aid in the real-time exploration of hazardous areas, robotically manipulate samples in real-time, and engage in precise measurements and experiments without the presence of crew in the immediate area. Furthermore, LLT can be particularly effective for studying ‘Special Regions’ – areas of astrobiological interest that might be adversely affected by forward contamination from humans or spacecraft contaminants during activities on Mars. LLT can also aid human exploration by addressing concerns about backward contamination that could impact mission details for returning Martian samples and crew back to Earth.This paper provides an overview of LLT operational considerations and findings from recent NASA analyses and workshops related to planetary protection and human missions beyond Earth orbit. The paper focuses primarily on three interrelated areas of Mars operations that are particularly relevant to the planetary protection and the search for life: Mars orbit-to-surface LLT activities; Crew-on-surface and drilling LLT; and Mars surface science laboratory LLT. The paper also discusses several additional mission implementation considerations and closes with information on key knowledge gaps identified as necessary for the advance of LLT for planetary protection and astrobiology purposes on future human missions to Mars.

Download Full-text

Multi-dimensional Taylor network modelling and optimal control of SISO nonlinear systems for tracking by output feedback

IMA Journal of Mathematical Control and Information ◽

10.1093/imamci/dnz020 ◽

2019 ◽

Vol 37 (3) ◽

pp. 699-717 ◽

Cited By ~ 1

Author(s):

Qi-Ming Sun ◽

Hong-Sen Yan

Keyword(s):

Neural Network ◽

Neural Networks ◽

Nonlinear Systems ◽

Real Time ◽

Output Feedback ◽

Tracking Error ◽

Real Time Control ◽

Network Modelling ◽

Single Input Single Output ◽

Time Control

Abstract In this paper, a multi-dimensional Taylor network (MTN) output feedback tracking control of nonlinear single-input single-output (SISO) systems in discrete-time form is studied. To date, neural networks are generally used to identify unknown nonlinear systems. However, the neuron of neural networks includes the exponential function, which contributes to the complexity of calculation, making the neural network control unable to meet the real-time requirements. In order to identify the controlled object whose model is unknown, the MTN, which requires only addition and multiplication, is utilized for successful real-time control of the SISO nonlinear system based on only its output feedback. Lyapunov analysis proves that output signals in the closed-loop system remain bounded and the tracking error converges to an arbitrarily small neighbourhood around the origin. In contrast to the back propagation (BP) neural network self-adaption reconstitution controller, the edge of the scheme is that the MTN optimal controller promises desirable response speed, robustness and real-time control.

Download Full-text