Robots need to know where they are to move around. This is hard for machines but easy for humans. Researchers at Queensland University of Technology have created a new system called LENS, which stands for Locational Encoding with Neuromorphic Systems. It helps robots navigate using very little energy, copying how the human brain works. This system uses neuromorphic computing, a technology that processes information like brain neurons. LENS needs less than 10 percent of the energy used by older systems, making it ideal for robots in places like search and rescue or space exploration, where power is limited.
The LENS system, described in a paper published in Science Robotics, includes a spiking neural network that mimics brain signals using electrical spikes. It also uses a special camera that only captures changes in movement or brightness, not full images. This is similar to how human eyes notice changes in a scene. The system runs on a low-power chip. Together, these parts allow a robot to recognize places along an 8-kilometer path while using very little storage, about 180 kilobytes, which is hundreds of times less than other systems.
A step toward energy-efficient robotics
This technology reduces energy use by up to 99 percent, letting robots work longer and travel farther on small power supplies. Unlike older neuromorphic systems, which were complex and hard to use, LENS is practical for real-world robots. The researchers designed special algorithms to process information like the brain does. These algorithms help the robot locate itself using only visual information, making navigation fast and efficient.
The system’s design opens new possibilities for robots that operate remotely, such as in underwater or space missions. By combining brain-inspired computing with advanced cameras and low-power chips, LENS sets a new standard for energy-efficient navigation. This work shows how technology can be both cutting-edge and practical, meeting the needs of users who rely on robots for challenging tasks. The study proves that brain-like systems can make robots more capable and sustainable for future applications.