Wind and obstacle motion affect honeybee flight strategies in cluttered environments

ABSTRACTBees often forage in habitats with cluttered vegetation and unpredictable winds. Navigating obstacles in wind presents a challenge that may be exacerbated by wind-induced motions of vegetation. Although wind-blown vegetation is common in natural habitats, we know little about how the strategies of bees for flying through clutter are affected by obstacle motion and wind. We filmed honeybees Apis mellifera flying through obstacles in a flight tunnel with still air, headwinds or tailwinds. We tested how their ground speeds and centering behavior (trajectory relative to the midline between obstacles) changed when obstacles were moving versus stationary, and how their approach strategies affected flight outcome (successful transit versus collision). We found that obstacle motion affects ground speed: bees flew slower when approaching moving versus stationary obstacles in still air but tended to fly faster when approaching moving obstacles in headwinds or tailwinds. Bees in still air reduced their chances of colliding with obstacles (whether moving or stationary) by reducing ground speed, whereas flight outcomes in wind were not associated with ground speed, but rather with improvement in centering behavior during the approach. We hypothesize that in challenging flight situations (e.g. navigating moving obstacles in wind), bees may speed up to reduce the number of wing collisions that occur if they pass too close to an obstacle. Our results show that wind and obstacle motion can interact to affect flight strategies in unexpected ways, suggesting that wind-blown vegetation may have important effects on foraging behaviors and flight performance of bees in natural habitats.

Counter a Drone in a Complex Neighborhood Area by Deep Reinforcement Learning

Counter-drone technology by using artificial intelligence (AI) is an emerging technology and it is rapidly developing. Considering the recent advances in AI, counter-drone systems with AI can be very accurate and efficient to fight against drones. The time required to engage with the target can be less than other methods based on human intervention, such as bringing down a malicious drone by a machine-gun. Also, AI can identify and classify the target with a high precision in order to prevent a false interdiction with the targeted object. We believe that counter-drone technology with AI will bring important advantages to the threats coming from some drones and will help the skies to become safer and more secure. In this study, a deep reinforcement learning (DRL) architecture is proposed to counter a drone with another drone, the learning drone, which will autonomously avoid all kind of obstacles inside a suburban neighborhood environment. The environment in a simulator that has stationary obstacles such as trees, cables, parked cars, and houses. In addition, another non-malicious third drone, acting as moving obstacle inside the environment was also included. In this way, the learning drone is trained to detect stationary and moving obstacles, and to counter and catch the target drone without crashing with any other obstacle inside the neighborhood. The learning drone has a front camera and it can capture continuously depth images. Every depth image is part of the state used in DRL architecture. There are also scalar state parameters such as velocities, distances to the target, distances to some defined geofences and track, and elevation angles. The state image and scalars are processed by a neural network that joints the two state parts into a unique flow. Moreover, transfer learning is tested by using the weights of the first full-trained model. With transfer learning, one of the best jump-starts achieved higher mean rewards (close to 35 more) at the beginning of training. Transfer learning also shows that the number of crashes during training can be reduced, with a total number of crashed episodes reduced by 65%, when all ground obstacles are included.

Dynamics of Swimmers in Fluids with Resistance

Micro-swimmers such as spermatozoa are able to efficiently navigate through viscous fluids that contain a sparse network of fibers or other macromolecules. We utilize the Brinkman equation to capture the fluid dynamics of sparse and stationary obstacles that are represented via a single resistance parameter. The method of regularized Brinkmanlets is utilized to solve for the fluid flow and motion of the swimmer in 2-dimensions when assuming the flagellum (tail) propagates a curvature wave. Extending previous studies, we investigate the dynamics of swimming when varying the resistance parameter, head or cell body radius, and preferred beat form parameters. For a single swimmer, we determine that increased swimming speed occurs for a smaller cell body radius and smaller fluid resistance. Progression of swimmers exhibits complex dynamics when considering hydrodynamic interactions; attraction of two swimmers is a robust phenomenon for smaller beat amplitude of the tail and smaller fluid resistance. Wall attraction is also observed, with a longer time scale of wall attraction with a larger resistance parameter.

Corridor navigation robot using Bond graph

In this research we are trying to find a mobile robot control method for corridor navigation and wall following in a partially known environment, when obstacles trajectory is unknown or information about it is incomplete, and the environment consists of stationary obstacles. Gathering information of the robot traveling is based on IR sensors. The aim of the robot is to select avoidance maneuvers to avoid collision with obstacles.

Four-Direction Search Scheme of Path Planning for Mobile Agents

SummaryThis paper presents a neural network-based four-direction search scheme of path planning for mobile agents, given a known environmental map with stationary obstacles. Firstly, the map collision energy is modeled for all the obstacles based on neural network. Secondly, for the shorted path-search purpose, the path energy is considered. Thirdly, to decrease the path-search time, a variable step-length is designed with respect to collision energy of the previous iteration path. Simulation results demonstrate that the variable step-length is effective and can decrease the iteration time substantially. Lastly, experimental results show that the mobile agent tracks the generated path well. Both the simulation and experiment results substantiate the feasibility and realizability of the presented scheme.

COLREGS-Constrained Real-time Path Planning for Autonomous Ships Using Modified Artificial Potential Fields

This paper presents a real-time and deterministic path planning method for autonomous ships or Unmanned Surface Vehicles (USV) in complex and dynamic navigation environments. A modified Artificial Potential Field (APF), which contains a new modified repulsion potential field function and the corresponding virtual forces, is developed to address the issue of Collision Avoidance (CA) with dynamic targets and static obstacles, including emergency situations. Appropriate functional and safety requirements are added in the corresponding virtual forces to ensure International Regulations for Preventing Collisions at Sea (COLREGS)-constrained behaviour for the own ship's CA actions. Simulations show that the method is fast, effective and deterministic for path planning in complex situations with multiple moving target ships and stationary obstacles and can account for the unpredictable strategies of other ships. The authors believe that automatic navigation systems operated without human interaction could benefit from the development of path planning algorithms.