Innovative robotic navigation inspired by brain functions improves efficiency and precision

A QUT research team has drawn inspiration from the brains of insects and animals for more energy-efficient robotic navigation.

Led by postdoctoral researcher Somayeh Hussaini, alongside Professor Michael Milford and Dr Tobias Fischer from the QUT Center for Robotics, the research, which was published In IEEE Transactions on Roboticsproposes a new place recognition algorithm using Spike Neural Networks (SNN).

“SNNs are artificial neural networks that mimic the way biological brains process information using brief, discrete signals, much like the way neurons in animal brains communicate,” Hussaini said.

“These networks are particularly well suited to neuromorphic hardware, specialized computing hardware that mimics biological neural systems, enabling faster processing and significantly reduced power consumption.”

Even though robotics has seen rapid advancements in recent years, modern robots still struggle to navigate and operate in complex and unfamiliar environments. They also often rely on AI-derived navigation systems whose training regimes require significant computing and energy resources.

“Animals are remarkably capable of navigating large, dynamic environments with astonishing efficiency and robustness,” Dr Fischer said.

“This work is a step toward the goal of biologically inspired navigation systems that could one day rival, or even surpass, today’s more conventional approaches.”

The system developed by the QUT team uses small neural network modules to recognize specific locations from images. These modules were combined into a package, a group of several state-of-the-art networks, to create a scalable navigation system capable of learning to navigate in large environments.

“Using image sequences instead of single images resulted in a 41% improvement in location recognition accuracy, allowing the system to adapt to changes in appearance over time and across different seasons and weather conditions,” Professor Milford said.

The system was successfully demonstrated on a resource-constrained robot, demonstrating that the approach is practical in real-world scenarios where energy efficiency is essential.

“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 areas such as space exploration and disaster recovery, where optimizing energy efficiency and reducing response times are essential,” said Hussaini. .