DNA computing is expanding its scope from mere DNA-based computing representation to developing integrated systems based on such computing. These integrated systems are called hybrid systems, and they are based on molecular computers, more precisely, molecular robots. This paper discusses molecular robots.
The authors examine the requirements of molecular computers in molecular robots, namely reactiveness, statefulness, hybridness, and persistency. They further find the components of a hybrid automaton, such as real-valued variables, discrete states (also called control modes), and discrete transitions (also called jumps). Furthermore, the authors describe the workflow of a hybrid automaton in the following three steps: identifying the inputs from the environment, designing the modes of the controller, and designing the transitions.
The authors describe the chemotaxis controller as an example hybrid controller implementing the following sections, input signal, the states of the controller, clock transitions, nonclock transitions, and domain-level implementation. The authors report two drawbacks in the existing DNA computing frameworks: “state transitions are not as sharp as expected for hybrid automata, and jump conditions require high Hill coefficients.”
This is an interesting read for nanotechnologists, professionals, and research scientists working in the areas of DNA computing, molecular computers, and robots. The description about using liposomes for jump transitions makes this paper worth reading. For future insights, the authors stress developing hybrid systems capable of implementing chemical reactions and physical phenomena in molecular robots. They further propose combining multiple approaches--whiplash machines, dynamic systems, and strand displacement reactions--while developing molecular robots.