About this Research Topic
With the explosive of communication technology, the multi-robot systems (MRS) have been widely used in search, rescue, industry and other fields. On the plus side, the homogeneous and heterogeneous systems have advantages of well fault-tolerant, robustness and strong ability to adapt dynamic environments. In spite of this, the task planning and collaboration control problems always exist in the large-scale pratical applications because single node only has limited capabilities of computation and communication and the control models have difficult to build to match real world applications. For example, for task allocation, the control model can include multiobjectives (e.g., minimum execution time and maximum completion rate) and multiconstraints, including limited energy, different node types, different task types and unknown environments. At present, there have been many methods for addressing the MRS control design and collaboration control problems. From the perspective of algorithm design, classical integer programming algorithm, markov decision algorithm and auction algorithm are usually employed. However, most of them still require human expert knowledge and high computation costs. Therefore, efficient automatic control for modelling and design is urgently needed.
As an enhanced machine learning (ML) by exploiting evolutionary computation (EC), evolutionary computation-based machine learning (ECML) integrates the advantages of both ML and EC to have the powerful potentials showing the excellent automatic control for addressing the modelling and design problems. ECML can adapt to the dynamic change of the number of nodes due to the adaptive ability so that the controlled modelling results are considered stable. Especially, ECML also has a strong search ability, which can decrease the computation cost of cluster nodes analytics to a large extent. Therefore, it is of great interest to investigate the role of ECML techniques in MRS.
This special issue focuses on ECML for multi-robot systems in terms of theoretical and practical issues. The purpose of this special issue is to bring together researchers, industry personnel, academicians and individuals working in these areas and to exchange novel ideas and the latest findings. The original papers are solicited on topics of interest that include, but are not limited to the following：
1. Multi-robot system modelling, optimization and design
2. Scalable ECML architecture for multi-robot systems
3. ECML for multi-objective optimization
4. ECML for high-dimensional system structure
5. ECML for hyperparameters control for heterogeneous multi-robot systems
6. ECML for task allocation
7. ECML for task scheduling
8. ECML for multi-robot systems
9. Integrative multi-robot controls of diverse, online, and offline environment
10. Large-scale and high-dimensional optimization for machine learning model
11. ECML-based multi-robot systems for unmanned aerial vehicle, business intelligence, finance, healthcare, bioinformatics, space exploration, agriculture, intelligent transportation, smart city, and other critical application areas
As an enhanced machine learning (ML) by exploiting evolutionary computation (EC), evolutionary computation-based machine learning (ECML) integrates the advantages of both ML and EC to have the powerful potentials showing the excellent automatic control for addressing the modelling and design problems. ECML can adapt to the dynamic change of the number of nodes due to the adaptive ability so that the controlled modelling results are considered stable. Especially, ECML also has a strong search ability, which can decrease the computation cost of cluster nodes analytics to a large extent. Therefore, it is of great interest to investigate the role of ECML techniques in MRS.
This special issue focuses on ECML for multi-robot systems in terms of theoretical and practical issues. The purpose of this special issue is to bring together researchers, industry personnel, academicians and individuals working in these areas and to exchange novel ideas and the latest findings. The original papers are solicited on topics of interest that include, but are not limited to the following：
1. Multi-robot system modelling, optimization and design
2. Scalable ECML architecture for multi-robot systems
3. ECML for multi-objective optimization
4. ECML for high-dimensional system structure
5. ECML for hyperparameters control for heterogeneous multi-robot systems
6. ECML for task allocation
7. ECML for task scheduling
8. ECML for multi-robot systems
9. Integrative multi-robot controls of diverse, online, and offline environment
10. Large-scale and high-dimensional optimization for machine learning model
11. ECML-based multi-robot systems for unmanned aerial vehicle, business intelligence, finance, healthcare, bioinformatics, space exploration, agriculture, intelligent transportation, smart city, and other critical application areas
Keywords: Evolutionary computation-based machine learning, Multi-robot system, Robotic control design, Bio-inspired control
Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.
