Controlling the motion of particles on a vibrating plate using dynamic acoustic fields

Abstract

Acoustic manipulation, a technique that moves objects by sound, has emerged as a promising method for handling of matter with a wide range of applications in biomedical research, microsystem assembly, lab-on-a-chip, and tissue engineering. Classical acoustic manipulation techniques operate by forming standing pressure waves and trapping the particles in the nodes or antinodes of the waves, enabling the formation of simple patterns of particles. During the last decade, a progress towards more dynamic devices has been initiated, resulting in the emergence of dynamic-field devices. Dynamic-field devices are able to move the acoustic traps, and accordingly the trapped objects, by dynamically reshaping the acoustic field. They enable complex manipulations such as moving biological organisms along predefined trajectories. Despite the remarkable achievements, the state-of-the-art acoustic manipulation methods face two major challenges: (1) The methods depend on the acoustic traps for manipulation, imposing clear functional limitations, e.g., to operate in the whole workspace, the device needs to create trapping points in that space; (2) Motion decoupling is challenging as the acoustic fields are global and when created, certain forces imposed by the shape of the acoustic field are applied to the particles and couple their motion. The suggested methods to solve these challenges typically demand a complex hardware with several, even hundreds, of transducers. First and foremost, this thesis introduces a new perspective on acoustic manipulation methods, which suggests motion control out of the acoustic traps. The idea has been applied to a vibrating plate in two environments, in air and underwater. It has major benefits compared to the state-of-the-art methods, where it considerably simplifies the hardware. For instance, this thesis shows that a single acoustic source can be used to simultaneously control the motion of up to six particles. Secondly, the thesis reports a novel method to control the motion of multiple objects independently and simultaneously inside a global field. It proposes employing a spatially highly nonlinear excitation field, but still global, for independent and simultaneous manipulation of multiple objects. The method allows complex operations on a vibrating plate in air and underwater, such as multi-particle manipulation on user-specific trajectories, pattern formation and transformation, and particle sorting. Finally, the thesis introduces a model-free control method based on reinforcement learning for dynamic-field devices. In this method, the controller does not need a prior knowledge of the acoustic field and learns the optimal control policy for each manipulation task by merely interacting with the acoustic field. The thesis reports the successful implementation of the method to a vibrating plate, allowing manipulation of single and multiple particles towards target locations.