Multi-robot systems are destined to revolutionize many industrial and research applications from warehouse management to search and rescue operations. They offer tremendous advantages over single robot systems since they can divide the workload and collaborate to achieve larger and more complex tasks. Mapping systems are frequently the backbone of those applications since such complex tasks usually rely on accurate representations of the environment from high-level features to detailed geometric information. In order to collaborate efficiently, multi-robot systems also require a shared situational awareness between the agents. In fact, having even a single agent with access to global localization sharing partial information with others can greatly benefit a robotic team. Additionally, the integration of robot planning is pivotal in orchestrating these collaborative efforts efficiently, ensuring optimal task allocation and navigation strategies that adhere to the shared goals and constraints of the operation.

Despite the massive progress in multi-robot localization, mapping, and planning in recent years, many challenges remain. Similar to single robot systems, the issues of long-term operations, robustness to failures, and resources management are prominent. New techniques have to address the limited storage and computation capabilities onboard robots, the time-critical nature of some applications, and/or the inability to operate in difficult environments. There is also the need for actionable maps with a deeper understanding of the environments whether it be through semantic labels, traversability analysis, or other useful information for mission planning. Robot planning further accentuates the requirement for these actionable insights to devise strategies that are both efficient and adaptable to changing conditions. On top of this, collaboratively mapping, localizing, and planning within a group of robots implies coordination and communication with limited bandwidth and changing topologies. Therefore, communication has to be restrained to a minimum, and it has to be resilient to failures and disruptions. In addition, collaborative techniques, due to their complexity and composite nature, are hard to reproduce and compare, so there is a need for standardized datasets and benchmarks to truly evaluate the progress of the field.

This Research Topic aims to collect publications presenting the latest state-of-the-art in the fields of localization, mapping, and robot planning applied to multi-robot systems. It welcomes Original Research contributions, Reviews, Perspectives, and Technology Reports related to, but not limited to, the following topics:
- Multi-robot Localization and/or Mapping in difficult environments
- Resources management for multi-robot localization, mapping, and planning
- Multi-robot state estimation
- Deep learning-enhanced multi-robot Localization and/or Mapping
- Active localization, mapping, and planning for multiple robots
- Multi-robot Localization and/or Mapping for efficient exploration
- Robot planning strategies for optimal task execution and resource allocation in multi-robot systems
- Multi-robot map merging and place recognition algorithms for multi-robot systems
- Multi-robot cameras, lidars, and other sensors datasets
- Novel applications of Multi-Robot Localization, Mapping, and Planning

Submit your idea

Impact

Frontiers in Robotics and AI

Submit your idea

Impact