Robotic control is a field of engineering with a clear demand for the efficient processing of spatial-temporal data, a category of task that is considered a core strength of Spiking Neural Networks (SNNs). However, the higher resource demands of most spiking neuron models means that use of conventional neural network architectures clashes with robotics’ demand for lower power and hardware overheads. A solution to this dichotomy can be found in the recently proposed Artificial Neural Microcircuit (ANM) paradigm, which proposes a methodology for the creation of compact yet highly capable spiking neuron based circuits [1], providing a more natural and desirable high-performing “event-driven” behaviour. This work reports on the development of a spiking neuron based control system, inspired by the Artificial Neural Microcircuit concept and demonstrated in embedded hardware, capable of random exploration and obstacle avoidance behaviours. Leveraging the Leaky Integrate-and-Fire (LIF) neuron model, the controller efficiently processes sensory data and actuates responses while maintaining a minimal number of neurons. Testing has demonstrated effective integration between hardware peripherals and software, with particular focus paid to the creation of a range of responses to specific obstacle scenarios.