Inspired by the locust's sense of smell, American scientists have developed a new biomimetic robotic sensing system that can be used for security applications such as detecting explosives...
A team of engineers at Washington University in St. Louis has developed a new biomimetic robotic sensing system inspired by the sense of smell of locusts, which could be used for security applications such as detecting explosives.
Baranidharan Raman, an associate professor of biomedical engineering in the Department of Engineering and Applied Sciences at the university, and with a $750,000 grant from the U.S. Naval Research Laboratory (ONR), developed a bio-hybrid nose based on this highly sensitive olfactory system.
Raman stated that biosensing systems are more complex than other similar engineered systems, including chemosensory systems responsible for olfactory senses. Although this olfactory sense is a primitive instinct, it still remains in many vertebrate and invertebrate species.
“Solutions that incorporate biology can be used to address problems such as non-invasive or chemical sensing, and the same design and operational principles are replicated everywhere,” Raman said. “Therefore, understanding the basic principles of olfactory processing is essential for biology-inspired engineering solutions.”
For years, and with funding from the ONR, Raman has been studying how the locust brain receives and processes sensory signals in a remarkably simple way. He and his team discovered that odors stimulate dynamic neural activity in the brain, allowing locusts to correctly identify specific odors, even when they are present alongside other odors. In other areas of research, the team has also found that locusts can be trained to recognize a particular odor successfully, even in complex situations, such as when it overlaps with other odors or in different background conditions.
“Why go through all that trouble? Why not just use a biological solution?” Raman points out. “Even the most advanced miniature chemical sensing devices have many sensors. On the other hand, if you look at an insect’s antenna—where its chemical sensors are located—there are millions of sensors of all types.”
The research team also plans to monitor the neural activity in the insects' brains as they move freely, exploring and decoding odors in their environment.
This method also requires low-power electronics to collect, record, and transmit data. Shantanu Chakrabartty, a professor of computer science and engineering and an expert in microelectronics at the Lab for Adaptive Integrated Microsystems, will collaborate with Raman on this part of the work.
This approach also requires low-power electronic components to collect, record, and transmit data. Shantanu Chakrabartty, a professor in the Department of Computer Science and Engineering at Washington University in St. Lewis, will collaborate with Raman to develop the components needed for this research.
The research team also plans to use locusts as a biomimetic robotic system for collecting samples via remote control. Srikanth Singamaneni, a multifunctional nanomaterials expert and assistant professor of materials science at the university, will develop a plasma "tattoo paper" made of biocompatible filaments that can be attached to the wings of locusts (generating a slight heat) and helps guide them to specific locations via remote control. Furthermore, the tattoo paper embedded with plasma nanostructures can also collect samples of nearby volatile organic compounds, allowing researchers to perform secondary analyses of the compounds' chemical composition using conventional methods.
“In many engineering applications, the canine olfactory system remains an advanced sensing system, including homeland security and medical diagnostics where sniffer dogs are still in use,” Raman said. “However, the time and difficulty required to train and tame these animals, coupled with the lack of robust decoding programs to extract chemically transmitted information from biological systems, pose significant challenges for broader applications.”
The difficulty and time required to train and regulate these animals, coupled with the lack of robust decoding procedures to extract information about chemical transmissions from biological systems, pose a significant challenge to the wider application of this approach. "We expect this research to develop and validate this locust-based chemical sensing pathway for further use in detecting explosives."
