|About the Book|
This dissertation proposes a novel data-driven two degree-of-freedom control scheme for life science automation. Particular attention is paid to the objective of designing a highly robust and intelligent system to achieve safety and reliability. TheMoreThis dissertation proposes a novel data-driven two degree-of-freedom control scheme for life science automation. Particular attention is paid to the objective of designing a highly robust and intelligent system to achieve safety and reliability. The two degree-of-freedom structure presents a combination of feedforward and feedback with guaranteed stability. The feedforward controller provides the basal energy to keep the system dynamics in the desired trajectory, while the feedback controller drives the system to the desired trajectory and guarantees the stability. The new approach improves on the inversion-based feedforward design to make finite-time transitions between stationary setpoints. Instead of using time to drive the smooth trajectory linking two terminal points, an intelligent event directly derived from the measurement is used. The integrated intelligent planning technique successfully avoids the large computational burden of real-time trajectory regeneration. Great potential in dealing with measurement noise and unexpected disturbances is demonstrated via applications in life science automation.-We explore two special cases in life science automation: drug delivery and gene delivery. Dynamics modeling on both micro/nanoscale systems are investigated via theoretical analysis and computer simulation. Many challenging nonlinear characteristics of these systems appeal to our interest, including hysteresis, multiple scales, bifurcation and slow response rates. The performance of the proposed data-driven two degree-of-freedom controller is superior to the existing methods in the literature.