Multi-robot task planning and collaboration are critical challenges in robotics. While Behavior Trees (BTs) have been established as a popular control architecture plannable for single robot, the development of effective multi-robot BT algorithms remains challenging due to complexity coordinating diverse action spaces. We propose Multi-Robot Tree Planning (MRBTP) algorithm, with theoretical guarantees both soundness completeness. MRBTP features cross-tree expansion coordinate heterogeneous...

Programming control systems for mobile robots is complicated and time-consuming, due to three aspects, i.e., the robot behavior coordination, distributed multi-robot cooperation software reusability. Subsumption model a robust architecture robots. ALLIANCE extends it multirobot systems, which fully distributed, fault-tolerant model. Robot operating system (ROS) provides lot of reusable modules. By combining above three, we propose framework named ALLIANCE-ROS developing cooperative with...

The performance of decentralized multi-agent systems tends to benefit from information sharing and its effective utilization. However, too much or unnecessary may hinder the due delay, instability additional overhead communications. Aiming a satisfiable coordination performance, one would prefer cost communications as less possible. In this paper, we propose an approach for improving utilization by integrating with prediction in planning. We present novel planning algorithm combining...

Behavior Trees (BTs) have attracted much attention in the robotics field recent years, which generalize existing control architectures and bring unique advantages for building robot systems. Automated synthesis of BTs can reduce human workload build behavior models complex tasks beyond ability design, but theoretical studies are almost missing methods because it is difficult to conduct formal analysis with classic BT representations. As a result, they may fail that actually solvable. This...

Abstract In this paper, we propose a distributed control approach for flocking and group maneuvering of nonholonomic agents, with constrained kinematic properties commonly found in practical systems, such as fixed‐wing unmanned aerial vehicles. Flocking agents differential drive kinematics is addressed by introducing virtual leader–follower mechanism into the Olfati‐Saber's algorithm, which originally proposed holonomic double integrator kinematics. Then, maneuverability flock achieved...

The agent-based subsumption model is widely acknowledged as the control systems for mobile robots. In this model, incremental layers can be stacked together by inhibitors and suppressors, which increasingly leads to complex coordinate behaviors. ROS (Robot Operating System) an open source robot software platform gradually becoming de facto standard applications. There are abundant reusable function units in ROS, coupled a distributed messaging mechanism. This paper describes template based...

Deep reinforcement learning is making advances in robotics with the platforms of realistic environment simulation. However, as shown this paper, simulation introduces vast time cost which bottleneck procedure. To solve problem generally, we propose a parallel platform follows master-slave principle and integrates programs multiple distributedrobot simulators. The intrinsically scalable requires no modification to existing serially designed environments or algorithms. Experimental results...

Multi-robot system has important application potential in disaster rescue and other dangerous scenarios, which task allocation is the basis for multi-robot cooperation to complete tasks. Due limited ability of individual robots, many complex scenarios requires coordination different types robot, e.g. robots search, communication, so on. At same time, communication often a wide range scenarios. This brings two challenges realistic environment: heterogeneous problem constraint problem. paper...

Abstract Deep reinforcement learning (DRL) has greatly improved the intelligence of AI in recent years and community proposed several common software to facilitate development DRL. However, robotics utility DRL is limited time-consuming due complexity various robot software. In this paper, we propose a engineering approach leveraging modularity development. The platform decouples environment into task, simulator hierarchical modules, which turn enables diverse generation using existing...

Multi-agent coordination tends to benefit from efficient communication, where cooperation often happens based on exchanging information about what the agents intend do, i.e. intention sharing. It becomes a key problem model by some proper abstraction. Currently, it is either too coarse such as final goals or fined primitive steps, which inefficient due lack of modularity and semantics. In this paper, we design novel multi-agent protocol subgoal intentions, defined probability distribution...

Unmanned Aerial Vehicles (UAVs) have been attracting more and attention in research education. Specifically, Swarm intelligence is a promising future technology of UAVs the frontier multi-agent system research. It has characteristics low individual cost, strong flexibility robustness, great potential many tasks. However, due to constraints conditions most current researches on large-scale swarm are carried out simulation environment. Building low-cost open-source software hardware platform...

The capability of a robot to perform tasks depends not only on precise motion control, but also well-suited body morphology. Adapting both morphology and control robots improve their task performance has been widely studied long-standing issue. While the bio-inspired bi-level optimization framework gained popularity in recent years, it suffers from high computation complexity due time-consuming inefficient learning process for each In fact, nature, besides adaptive intelligent brain, animals...

An agent's intelligence tends to benefit from the co-adaption between body morphology and behavior policy, aka. embodied intelligence. To create an agent, joint optimization of becomes key problem. solve that, a popular approach is use search-based or learning-based heuristics traverse morphological space, optimize policy for each evaluate their fitness. However, prior works often ignore relations components modular behaviors, e.g. mechanical foot can be used walk, run, kick, etc. reduce...

Unmanned aerial vehicles (UAVs) have been attracting more and attention in the research industry field. Aerial search is a common mission intrinsically fit for UAVs, e.g. disaster rescue, remote sensing environmental monitoring. With improvement of UAV hardware software, UAVs tend to achieve better autonomy accomplish complex tasks. However, current usually hardcoded, which limits their adaptability, robustness realistic scenarios. In this paper, we propose address problem by leveraging...