In this thesis, the effects of process parameters on the resulting feature morphology and dimensions within line length scales and micro-fluidic devices is presented. Positioning stages, laser systems, and autonomous control systems were developed and designed for the machining of micro-channels on glass sheet and inside polycarbonate and PMMA samples. The developed real time closed loop control system was set-up via reconfigurable I/O Field-Programmable Gate Array (FPGA). In-depth analyses of the positional performance of the developed Nd:YVO4, and Nd:YAG laser systems were carried out. The results of these analyses indicated that the developed 3D translation stage of the Nd:YVO4 laser system is better with accuracy and repeatability values less than 65 µm for all the three axes. In particular, CO2 (1.5 kW, 10.6 µm) and Nd:YVO4 (2.5 W, 1.064 µm) laser systems were investigated experimentally and through system models in order to better understand the effects of laser and motion parameters on the process control. Predictive models, that relate the laser machining parameters (laser power; P, pulse repetition frequency; PRF, and sample translation speed; U) to the geometry and cost of the produced micro-channels, were developed. Detailed designs of experiments (DoE) were conducted and results from developed predictive models based on Artificial Neural Network (ANN) and Response Surface Methodology (RSM) techniques were compared with the actual results. Statistical estimators were used to evaluate these models and compare their predictive and generalization ability. Results showed that although the ANN models provided the highest prediction accuracy, both RSM and ANN modelling techniques could be utilised as effective predictive tools for resultant laser micro-machined dimensions and selection of laser micromachining parameters.