Research harnesses brain function for robot navigation

The team, which was led by postdoctoral research fellow Somayeh Hussaini, along with Professor Michael Milford and Dr Tobias Fischer of the QUT Centre for Robotics and supported by chip manufacturer Intel, proposed a novel place recognition algorithm using Spiking Neural Networks (SNNs).

“SNNs are artificial neural networks that mimic how biological brains process information using brief, discrete signals, much like how neurons in animal brains communicate,” Hussaini explained. “These networks are particularly well-suited for neuromorphic hardware—specialised computer hardware that mimics biological neural systems—enabling faster processing and significantly reduced energy consumption.”

Although robotics as an industry has undergone considerable progress, one area that continues to face challenges is in navigation. AI-derived navigation systems, which are commonly used, have significant computational and energy requirements.

The system, which the QUt team developed, uses small neural network modules to recognise specific places from images. The modules were combined into an ensemble, a group of multiple spiking networks, to form a scalable navigation system capable of learning to navigate in large environments.

“Animals are remarkably adept at navigating large, dynamic environments with amazing efficiency and robustness,” said Dr Fischer. “This work is a step towards the goal of biologically inspired navigation systems that could one day compete with or even surpass today’s more conventional approaches.”

“Using sequences of images instead of single images enabled an improvement of 41% in place [of] recognition accuracy, allowing the system to adapt to appearance changes over time and across different seasons and weather conditions,” Professor Milford said.

The system was demonstrated using a resource-constrained robot to provide a proof of concept that the system is valuable in real-world scenarios.

“This work can help pave the way for more efficient and reliable navigation systems for autonomous robots in energy-constrained environments. Particularly exciting opportunities include domains like space exploration and disaster recovery, where optimising energy efficiency and reducing response times are critical,” Hussaini concluded.