Decentralized Coordination of a Multi-UAV System for Spatial Planar Shape Formations.

Motivated by feedback from firefighters in Normandy, this work aims to provide a simple technique for a set of identical drones to collectively describe an arbitrary planar virtual shape in a 3D space in a decentralized manner. The original problem involved surrounding a toxic cloud to monitor its composition and short-term evolution. In the present work, the pattern is described using Fourier descriptors, a convenient mathematical formulation for that purpose. Starting from a reference point, which can be the center of a fire, Fourier descriptors allow for more precise description of a shape as the number of harmonics increases. This pattern needs to be evenly occupied by the fleet of drones under consideration. To optimize the overall view, the drones must be evenly distributed angularly along the shape. The proposed method enables virtual planar shape description, decentralized bearing angle assignment, drone movement from takeoff positions to locations along the shape, and collision avoidance. Furthermore, the method allows for the number of drones to change during the mission. The method has been tested both in simulation, through emulation, and in outdoor experiments with real drones. The obtained results demonstrate that the method is applicable in real-world contexts.

Decentralized Policy Coordination in Mobile Sensing with Consensual Communication.

In a typical mobile-sensing scenario, multiple autonomous vehicles cooperatively navigate to maximize the spatial-temporal coverage of the environment. However, as each vehicle can only make decentralized navigation decisions based on limited local observations, it is still a critical challenge to coordinate the vehicles for cooperation in an open, dynamic environment. In this paper, we propose a novel framework that incorporates consensual communication in multi-agent reinforcement learning for cooperative mobile sensing. At each step, the vehicles first learn to communicate with each other, and then, based on the received messages from others, navigate. Through communication, the decentralized vehicles can share information to break through the dilemma of local observation. Moreover, we utilize mutual information as a regularizer to promote consensus among the vehicles. The mutual information can enforce positive correlation between the navigation policy and the communication message, and therefore implicitly coordinate the decentralized policies. The convergence of this regularized algorithm can be proved theoretically under certain mild assumptions. In the experiments, we show that our algorithm is scalable and can converge very fast during training phase. It also outperforms other baselines significantly in the execution phase. The results validate that consensual communication plays very important role in coordinating the behaviors of decentralized vehicles.

Effects of load mass and size on cooperative transport in ants over multiple transport challenges.

Some ant species cooperatively transport a wide range of extremely large, heavy food objects of various shapes and materials. While previous studies have examined how object mass and size affect the recruitment of additional workers, less is understood about how these attributes affect the rest of the transport process. Using artificial baits with independently varying mass and size, we reveal their effects on cooperative transport in Paratrechina longicornis across two transport challenges: movement initiation and obstacle navigation. As expected, object mass was tightly correlated with number of porters as workers adjust group size to the task. Mass affected performance similarly across the two challenges, with groups carrying heavy objects having lower performance. Yet, object size had differing effects depending on the challenge. While larger objects led to reduced performance during movement initiation - groups took longer to start moving these objects and had lower velocities - there was no evidence for this during obstacle navigation, and the opposite pattern was weakly supported. If a group struggles to start moving an object, it does not necessarily predict difficulty navigating around obstacles; groups should persist in trying to move 'difficult' objects, which may be easier to transport later in the process. Additionally, groups hitting obstacles were not substantially disrupted, and started moving again sooner than at the start, despite the nest direction being blocked. Paratrechina longicornis transport groups never failed, performing well at both challenges while carrying widely varying objects, and even transported a bait weighing 1900 times the mass of an individual.

Decentralized 3D Collision Avoidance for Multiple UAVs in Outdoor Environments.

The use of multiple aerial vehicles for autonomous missions is turning into commonplace. In many of these applications, the Unmanned Aerial Vehicles (UAVs) have to cooperate and navigate in a shared airspace, becoming 3D collision avoidance a relevant issue. Outdoor scenarios impose additional challenges: (i) accurate positioning systems are costly; (ii) communication can be unreliable or delayed; and (iii) external conditions like wind gusts affect UAVs' maneuverability. In this paper, we present 3D-SWAP, a decentralized algorithm for 3D collision avoidance with multiple UAVs. 3D-SWAP operates reactively without high computational requirements and allows UAVs to integrate measurements from their local sensors with positions of other teammates within communication range. We tested 3D-SWAP with our team of custom-designed UAVs. First, we used a Software-In-The-Loop simulator for system integration and evaluation. Second, we run field experiments with up to three UAVs in an outdoor scenario with uncontrolled conditions (i.e., noisy positioning systems, wind gusts, etc). We report our results and our procedures for this field experimentation.

Collective strategy for obstacle navigation during cooperative transport by ants.

Group cohesion and consensus have primarily been studied in the context of discrete decisions, but some group tasks require making serial decisions that build on one another. We examine such collective problem solving by studying obstacle navigation during cooperative transport in ants. In cooperative transport, ants work together to move a large object back to their nest. We blocked cooperative transport groups of Paratrechina longicornis with obstacles of varying complexity, analyzing groups' trajectories to infer what kind of strategy the ants employed. Simple strategies require little information, but more challenging, robust strategies succeed with a wider range of obstacles. We found that transport groups use a stochastic strategy that leads to efficient navigation around simple obstacles, and still succeeds at difficult obstacles. While groups navigating obstacles preferentially move directly toward the nest, they change their behavior over time; the longer the ants are obstructed, the more likely they are to move away from the nest. This increases the chance of finding a path around the obstacle. Groups rapidly changed directions and rarely stalled during navigation, indicating that these ants maintain consensus even when the nest direction is blocked. Although some decisions were aided by the arrival of new ants, at many key points, direction changes were initiated within the group, with no apparent external cause. This ant species is highly effective at navigating complex environments, and implements a flexible strategy that works for both simple and more complex obstacles.

Multi-robot replication of ant collective towing behaviours.

In this work, teams of small mobile robots are used to test hypotheses about cooperative transport by ants. This study attempts to explain a decrease in steady-state transport speed with increasing team size that was previously observed in the ant Novomessor cockerelli. Two models of one-dimensional collective towing are compared: one in which transporters with different maximum speeds pull the payload with continuous, variable forces and another in which transporters with identical speeds pull with intermittent, unsynchronized forces. A statistical analysis of ant data supports the hypothesis that ants behave according to the first model, in which the steady-state transport speed is the maximum speed of the slowest teammate. By contrast, the ant data are not consistent with the second model, which predicts constant speed regardless of team size. To verify these predictions, the ant behaviours in each model are translated into decentralized controllers and implemented on teams of two to four robots. The controller for the first model incorporates a real-time reinforcement learning algorithm that successfully reproduces the observed relationship between ant team size and transport speed. The controller for the second model yields the predicted invariance of transport speed with team size. These results show the value of robotic swarms for testing mechanistic hypotheses about biological collectives.