Neuro-robots utilize neural structures and mechanisms inspired by living creatures, especially insects, to achieve efficient sensing and reaction to the dynamic world. Thanks to the low computing power requirements, real-time capability, robustness, and strong interpretability of insect-inspired perception models, neuro-robots have demonstrated reliable performances in various robotics research domains, e.g.in micro-robots, swarm robots, bionic robotics, service robots, and unmanned ground/aerial systems. This shows the great potential of employing biological likelihood neuro-structures in solving real-world problems, especially for those micro-robots that need to work together in fleets or in swarms to accomplish missions beyond an individual’s capability.
This proposed special issue aims to promote an understanding of how the biologically inspired perception systems can be applied to micro neuro-robots for various applications. To this end, research in the following areas will be collected. One is the perceptual neural network model, especially the neurobiological model, algorithm, and structure of visual cue extraction, which should be discussed first. By directly encoding and abstracting motion information, one can quickly discriminate and respond to dynamic environmental stimuli. Then, the key challenges at the application level should be addressed, such as complexity issues and real-time optimization of algorithms, sparse representation of neuro-models, data set construction and learning methods, the collaboration of multi-agent systems, and the swarm intelligence. Finally, the value of neuro-robots in typical application scenarios should be further demonstrated, such as ruin search and rescue, path planning, rehabilitation treatment, elder and disability assistance, smart transportation systems, surveillance and patrol, manufacturing, and logistics, etc.
The development and application of neuro-robots is a bridge that combines exploration of the underlying mechanisms, in highly efficient biological neuro-structures, and solutions to the bottleneck of processing a large number of redundant information in the vast majority of artificial intelligence applications. The neuro-robot approach is a typical method of interdisciplinary collaboration, which also projects great positive feedback to biological research. Relevant topics include, but are not limited to:
- Neurorobotic visual perception
- Insect-inspired collision detection
- Neuro-inspired sensory systems
- Neurorobotic control strategies
- Visual motion cues in encoding and decoding
- Mathematical methodologies for neuromorphic modeling
- Neural network models optimization and validation
- Neuro-robots in swarm behavior studies
- Neuro-robots in real-world applications
Neuro-robots utilize neural structures and mechanisms inspired by living creatures, especially insects, to achieve efficient sensing and reaction to the dynamic world. Thanks to the low computing power requirements, real-time capability, robustness, and strong interpretability of insect-inspired perception models, neuro-robots have demonstrated reliable performances in various robotics research domains, e.g.in micro-robots, swarm robots, bionic robotics, service robots, and unmanned ground/aerial systems. This shows the great potential of employing biological likelihood neuro-structures in solving real-world problems, especially for those micro-robots that need to work together in fleets or in swarms to accomplish missions beyond an individual’s capability.
This proposed special issue aims to promote an understanding of how the biologically inspired perception systems can be applied to micro neuro-robots for various applications. To this end, research in the following areas will be collected. One is the perceptual neural network model, especially the neurobiological model, algorithm, and structure of visual cue extraction, which should be discussed first. By directly encoding and abstracting motion information, one can quickly discriminate and respond to dynamic environmental stimuli. Then, the key challenges at the application level should be addressed, such as complexity issues and real-time optimization of algorithms, sparse representation of neuro-models, data set construction and learning methods, the collaboration of multi-agent systems, and the swarm intelligence. Finally, the value of neuro-robots in typical application scenarios should be further demonstrated, such as ruin search and rescue, path planning, rehabilitation treatment, elder and disability assistance, smart transportation systems, surveillance and patrol, manufacturing, and logistics, etc.
The development and application of neuro-robots is a bridge that combines exploration of the underlying mechanisms, in highly efficient biological neuro-structures, and solutions to the bottleneck of processing a large number of redundant information in the vast majority of artificial intelligence applications. The neuro-robot approach is a typical method of interdisciplinary collaboration, which also projects great positive feedback to biological research. Relevant topics include, but are not limited to:
- Neurorobotic visual perception
- Insect-inspired collision detection
- Neuro-inspired sensory systems
- Neurorobotic control strategies
- Visual motion cues in encoding and decoding
- Mathematical methodologies for neuromorphic modeling
- Neural network models optimization and validation
- Neuro-robots in swarm behavior studies
- Neuro-robots in real-world applications