The UK’s University of Lincoln is leading an international research project that aims to create a pioneering new collision avoidance system to enhance the safety of driverless vehicles, taking its inspiration from swarming insects.
Funded by a €1.8m (US$2.2m) grant from the European Union’s Horizon 2020 research and innovation program, the project will develop a miniature, trustworthy collision detection sensor system that could drastically improve the safety of autonomous vehicles (AVs).
The ULTRACEPT (Ultra-layered perception with brain-inspired information processing for vehicle collision avoidance) project will develop a new microchip for AVs that aims to make them much safer using ‘swarm intelligence’.
Many AV developers have found that the current approaches for vehicle collision detection are largely ineffective in terms of reliability, cost, energy consumption and size, with issues encountered including:
• Radar is too sensitive to metallic material;
• GPS-based methods face difficulties in cities with high buildings;
• Vehicle-to-vehicle (V2V) communication cannot detect pedestrians or any unconnected objects;
• Normal vision sensors cannot cope with fog, rain or dim light conditions at night.
The ULTRACEPT researchers hope to develop a system that will overcome all of these issues.
The new ULTRACEPT sensor will be inspired by the rapid reactions of insects, incorporating near-range collision detection technology, long-range hazard perception, and thermal-based collision detection tools. This will ensure that it works day and night, and can quickly adapt to unexpected hazards and different conditions, such as sudden weather changes or driving in and out of tunnels.
The project brings together experts from universities in the UK, Germany, China, Japan, Malaysia, and Argentina, to create a robust, low-cost, and energy-efficient collision detection and avoidance system that will offer a capability which is beyond AVs currently in development.
The project involves a world-class research team with specialist knowledge, including experts in hardware and software systems and robotics, invertebrate visual neuroscientists, invertebrate vision modelers, mixed-signal chip designers, robotics platform providers, and brain-inspired pattern recognition.
“Biology provides a rich source of inspiration for artificial visual systems for collision detection and avoidance. For example, locusts, with a compact visual brain, can fly for hundreds of miles in dense swarms free of collision; the praying mantis can monitor tiny moving prey with the help of specialized visual neurons; and nocturnal insects successfully forage in the forest at night without collision,” explained ULTRACEPT project leader Professor Shigang Yue (above), from the University of Lincoln.
“These naturally evolved vision systems provide ideal models to develop an artificial system for collision detection and avoidance, and we hope that in the future, each vehicle, with or without a driver, will be well equipped with an innovative sensor to navigate as effectively as animals do.”
