Cloud-based computing infrastructure with enormous storage capability, coupled with new tools for remotely accessing, analysing and visualising data, is developing at an exponential rate in terms of the variety and scale of available information. The ‘sensor web’ is an area of particular interest, with many major industries increasingly looking at ways to apply sensor network technologies to develop new products and services. Sensor networks require a scalable model, in which the basic building block, the sensor (or sensing platform) is very low cost to buy, requires no maintenance, and has a very long lifetime (Years). Understandably, attention has therefore focused on sensors that meet these requirements, like thermistors, photodiodes, movement and vibration sensors, and imaging systems. Despite the obvious tremendous value of integrating molecular sensing capabilities into this concept via networked chemical sensors and biosensors, there are no examples of large-scale deployments of such devices. There are a number of reasons why this is so including capital cost, maintenance costs, reliability issues, and the hostile environments to which these devices are typically exposed. These issues are obviously inter-related, and are linked to the degradation of sensitive sensor surfaces that occurs upon deployment. This changes the device response characteristics, which in turn requires integration of calibration procedures into devices. Conventional approaches to liquid handling within chemical sensor platforms drives up the price and complexity, and requires larger form factors for reagent and waste storage. The integration of new concepts for liquid control using fully integrated photo-actuated polymer valves offers a way forward, if issues related to reliability and response characteristics can be solved. These circulation systems are more biomimetic in nature than traditional microfluidic platforms based on silicon micromachining. The introduction of mobile micro/nano-vehicles capable of performing various functions (stimuli-responsive control of movement, location, specific target binding, communications and controlled release) offer exciting additional capabilities based on biomimetic principles. In this paper, I will present recent results based on photo-switchable gels and microvehicles that suggest we are moving closer to the realisation of efficient yet sophisticated low-cost microfluidic platforms that could provide the basis for future autonomous chemo/bio-sensing technologies.