The recent advancement of Deep Reinforcement Learning (DRL) contributed to robotics by allowing automatic controller design. The automatic controller design is a crucial approach for designing swarm robotic systems, which require more complex controllers than a single robot system to lead a desired collective behaviour. Although the DRL-based controller design method showed its effectiveness for swarm robotic systems, the reliance on the central training server is a critical problem in real-world environments where robot-server communication is unstable or limited. We propose a novel Federated Learning (FL) based DRL training strategy (FLDDPG) for use in swarm robotic applications. Through the comparison with baseline strategies under a limited communication bandwidth scenario, it is shown that the FLDDPG method resulted in higher robustness and generalisation ability into a different environment and real robots, while the baseline strategies suffer from the limitation of communication bandwidth. This result suggests that the proposed method can benefit swarm robotic systems operating in environments with limited communication bandwidth, e.g., in high-radiation, underwater, or subterranean environments.

Digital and mechatronic methods, paired with artificial intelligence and machine learning, are transformative technologies in behavioral science and biology. The central element of the most important pollinator species—honey bees—is the colony’s queen. Because honey bee self-regulation is complex and studying queens in their natural colony context is difficult, the behavioral strategies of these organisms have not been widely studied. We created an autonomous robotic observation and behavioral analysis system aimed at continuous observation of the queen and her interactions with worker bees and comb cells, generating behavioral datasets of exceptional length and quality. Key behavioral metrics of the queen and her social embedding within the colony were gathered using our robotic system. Data were collected continuously for 24 hours a day over a period of 30 days, demonstrating our system’s capability to extract key behavioral metrics at microscopic, mesoscopic, and macroscopic system levels. Additionally, interactions among the queen, worker bees, and brood were observed and quantified. Long-term continuous observations performed by the robot yielded large amounts of high-definition video data that are beyond the observation capabilities of humans or stationary cameras. Our robotic system can enable a deeper understanding of the innermost mechanisms of honey bees’ swarm-intelligent self-regulation. Moreover, it offers the possibility to study other social insect colonies, biocoenoses, and ecosystems in an automated manner. Social insects are keystone species in all terrestrial ecosystems; thus, developing a better understanding of their behaviors will be invaluable for the protection and even the restoration of our fragile ecosystems globally.

When a mobile robot is asked to navigate intelligently in an environment, it needs to estimate its own and the environment's state. One of the popular methods for robot state and position estimation is particle filtering (PF). Visual Teach and Repeat (VT&R) is a type of navigation that uses a camera to navigate the robot along the previously traversed path. Particle filters are usually used in VT&R to fuse data from odometry and camera to estimate the distance traveled along the path. However, in VT&R, there are other valuable states that the robot can benefit from, especially when moving through changing environments. We propose a multidimensional particle filter to estimate the robot state in VT&R navigation. Apart from the traveled distance, our particle filter estimates lateral and heading deviation from the taught path as well as the current appearance of the environment. This appearance is estimated using maps created across various environmental conditions recorded during the previous traversals. The joint state estimation is based on contrastive neural network architecture, allowing self-supervised learning. This architecture can process multiple images in parallel, alleviating the potential overhead caused by computing the particle filter over the maps simultaneously. We conducted experiments to show that the joint robot/environment state estimation improves navigation accuracy and robustness in a continual mapping setup. Unlike the other frameworks, which treat the robot position and environment appearance separately, our PF represents them as one multidimensional state, resulting in a more general uncertainty model for VT&R.