We deal with the problem of incorporating the most recent information in the system which may be crucial to update the system state, while maintaining resilience to faulty measurements. For this purpose, we propose and analyze a consensus protocol biased towards the latest information in the system. The convergence to consensus of the system is analyzed and two main approaches are introduced with stability ranges provided from sufficient conditions. A detailed analysis of weight-changing that achieves bias towards novelty in the system is provided. Numerical simulations demonstrate that final value and convergence rate of the system can be controlled. Finally, we provide system demonstration in a simple use-case example.

This paper presents a method for extending decision-making policies in active perception tasks for multi-agent systems within partially observable environments. Multiple agents obtain their policies by training in an environment of a certain size. Those policies are then used in the environments larger in size, that are divided into sub-environments of size similar (or smaller) to one that the agents were trained in. Learned policies are adapted accordingly by proposed Action-Space Reduced Policy (ASRP). By leveraging Multi-Agent Reinforcement Learning (MARL) within a POMDP framework, agents can use their learned policies across environments of differing complexity without requiring retraining. The consensus mechanism allows agents to maintain a common belief state, supporting collaborative decision-making based on observations from all agents involved. Validation of the method is conducted on the scenario of multi-agent exploration missions, demonstrating the use of extended policies and enhanced perception accuracy. Simulation results indicate expected success rates and decision-making times, regardless of the environment’s dimensionality. Potential applications for scalable, multi-agent perception systems are discussed, along with directions for future research.

This paper proposes a unified dynamic model for aerial robots which encompasses all known actuation principles including tilting propellers and centroid variation methods such as moving masses or robotic manipulators. Of course, one can envision a wide variety of vehicles, all with different combinations of actuation principles. Therefore, a generalized modeling methodology for such vehicles is developed and presented in this paper. The modeling approach is verified through a comparative analysis of MATLAB simulations and laboratory experiments with a tilting propeller aerial manipulator vehicle, named Toucan. Finally, in order to fully explore and exploit the capabilities of the designed aerial manipulator, contact-based experiments with force tracking are performed using the proposed adaptive impedance controller. To demonstrate the advantages of such a vehicle, the adaptive impedance control method is used to generate position and orientation commands and achieve end-effector force tracking on a flat surface while the aerial manipulator maintains a neutral attitude. This greatly increases the stability and safety of contact-based operations, as the vehicle does not require changes in attitude to meet the force requirements. Note to Practitioners—This paper was motivated by a surge of different configurations of aerial vehicles emerging. A modeling method for generalized aerial vehicles is presented to determine which vehicle configurations are best suited for a specific application. This approach considers vehicles with additional actuators such as tilting propellers and robotic manipulators. The effectiveness of the proposed method is validated on a custom-designed quadrotor with four moving masses emulating centroid variation of a generic robotic manipulator. This original design is augmented with tilting propellers, yielding a truly generalized experimental platform. The generalized modeling algorithm and the experimental platform are tested both in free flight and in a scenario that demonstrates applying the desired force to a flat surface. Overall, we showcase the proposed generalized modeling is a useful tool for the rapid prototyping and design of aerial manipulators, paving the way for more advanced and capable systems in the future. With this approach, practitioners can gain a better understanding of the potential and limitations of their aerial vehicles, and ultimately improve their performance and effectiveness in specific applications.

We propose a new approach to visual perception for relative localization of agents within large-scale swarms of Unmanned Aerial Vehicles (UAVs). Inspired by biological perception utilized by schools of sardines, swarms of bees, and other large groups of animals capable of moving in a decentralized yet coherent manner, our method does not rely on detecting individual neighbors by each agent and estimating their relative position, but rather we propose to regress a neighbor density over distance. This allows for a more accurate distance estimation as well as better scalability with respect to the number of neighbors. Additionally, a novel swarm control algorithm is proposed to make it compatible with the new relative localization method. We provide a thorough evaluation of the presented methods and demonstrate that the regressing approach to distance estimation is more robust to varying relative pose of the targets and that it is suitable to be used as the main source of relative localization for swarm stabilization.

This paper presents a manipulator-equipped unmanned aerial vehicle (UAV) performing contact-based surface inspection. Using stereo camera information, a contact point is determined and an approach path is generated to ensure an autonomous workflow. The proposed inspection method uses an impedance controller to regulate the force applied by the end-effector to the contact surface. Due to the often adverse conditions in the real world, the aerial manipulator may fail to achieve contact after approaching the surface. Therefore, the impedance controller is coupled with an online stiffness estimation algorithm to increase the overall robustness of the inspection system and achieve stable surface contact. Finally, simulations in the Gazebo environment and experiments in the real world are performed to show the feasibility of this approach.

In this paper, we present a method for depth control of one degree of freedom (1DOF) underwater robotic platform aMussel, based on the measurements from the ambient light sensor. Since ambient light values change during the day and depend on the weather conditions, references for the controller are acquired from other aMussel holding depth using pressure sensor based controller. Control inputs are transmitted using acoustic communication.

This paper demonstrates the experimental validation of canonical peg-in-hole manipulation task using an aerial robot. The robot consists of a multirotor platform equipped with a dual arm multi degree of freedom manipulator. The paper discusses the introduced kinematic constraints which make sure the robot holds a bolt with both arms. We build our peg-in-hole approach using impedance control which is the foundation of compliant interaction with the environment. We utilize a finite state automaton to plan a multi stage strategy which relies on tactile perception in order to pin point the target. Finally, the whole body locomotion is considered, meaning both the degrees of freedom of multirotor base and the dual arm manipulator are considered.

In this paper, we present selected benchmark aerial manipulation tasks using an aerial vehicle endowed with multi-degree of freedom manipulators. The proposed tasks analyze environmental coupling and are broken into three general categories: momentary, loose, and strong coupling. A classical control structure is derived, tuned, and verified through experiments, conducted for benchmarking purposes to include pick-and-place, insertion, and valve-turning tasks. Although other nonlinear controllers may prove more effective, the classical control approach has been selected in order to analyze contact stability and provide benchmark results for future reference. An analysis of system stability is conducted and implemented into the controller. A vision-based high-level controller fuses motion tracking data in order to provide control of both the aircraft and the manipulators, allowing the system to become coupled to the environment and perform the required operation. We present recent results validating our framework using the proposed aircraft-arm system.

In this paper we present a study of a robotic system that consists of an unmanned aerial vehicle equipped with a pair of manipulator arms (MMUAV), and unmanned ground vehicles (UGVs). The envisioned application scenario includes autonomous packet transportation, where MMUAV is used for picking/placing packets, while both MMUAV and UGV can be used for packet transportation, with different energy consumption profiles. We propose a reactive method for decentralized task planning and coordination of robots using hierarchical task decomposition based on TÆMS framework. Our approach takes into account low-level motion-planning aspects of the system as well as high-level mission specification, making this a multi-layered system. For low-level planning we use sampling-based planner combined with obstacle-free trajectory generation. Methods are verified in simulations and on an experimental testbed, using 3D Robotics quadcopter and Pioneer 3DX mobile robots with the results showing stability and robustness of the presented methods.

In this paper, we present an algorithm for decentralized control of multiple automated guided vehicles performing transportation tasks within industrial and warehousing environments. By running on each vehicle in the system, the algorithm provides vehicles with capabilities for autonomous path planning and motion co-ordination. The path planning part of the algorithm implements a free-ranging motion scheme by determining the shortest feasible paths considering nonholonomic vehicle constraints. The motion co-ordination part of the algorithm ensures safe vehicle motions by reliable detection and resolution of different conflict situations with other vehicles in the shared workspace. Conflict resolution is based on a vehicle priority scheme and results in temporary stopping or removal of the lower priority vehicles taking part in the conflict. Removal action is always performed within the vehicle's private zone, i.e., the pre-allocated local region of the workspace surrounding the vehicle. By encoding information on the vehicle size and its kinematic constraints, the introduced private zone mechanism provides the necessary physical space required for successful execution of every removal action. We also analyze the stability of the presented algorithm and discuss its deadlock-free and livelock-free properties. Algorithm performance has been validated by simulation using ten vehicles and experimentally on two different setups-a laboratory setup comprising five Pioneer 3DX vehicles and by two state-of-the-art autonomous forklifts in industrial-like operating conditions.

This paper presents the results of our efforts to build a heterogeneous robotic system capable of executing complex disaster response and recovery tasks. We aim to explore high level task scheduling and mission planning algorithms that enable various types of robots to cooperate together, utilizing each others strengths to yield a symbiotic robotic system. In the proposed scenario, a ground vehicle and an aerial robot work together to close a valve in a disaster stricken industrial environment. To that end we use TAEMS framework in order to specify interrelationships between mission subtasks and develop an effective scheduling and coordination mechanism, inspired by Generalized Partial Global Planning. We present simulation results with two different outcomes that show cooperative capabilities of the system.

In this paper we propose an algorithm for decentralized control of Automated Guided Vehicles (AGVs) operating in automated warehouse environments. The motion planning part of the algorithm provides vehicles with capabilities for autonomous motion planning considering nonholonomic vehicle constraints and collision-free path execution. The decision making part of the algorithm ensures safe vehicle motions and reliable conflict situation resolution. The proposed control algorithm also prevents occurrence of deadlock and livelock situations. Stability of the algorithm has been proven by its analysis based on automata theory, while its performance has been validated by simulation on a system comprising twenty vehicles as well as experimentally on a laboratory setup comprising five Pioneer 3DX vehicles.

In this paper, we present a control system for a fully autonomous material handling facility. The scenario we are considering is motivated by the 2011 IEEE Virtual Manufacturing Automation Challenge (VMAC). It consists of multiple autonomously guided vehicles (AGVs), transporting pallets of various goods between several input and output locations, through an unstructured warehouse environment. Only a map of the warehouse and a pallet delivery list are provided a priori. Pallets must be delivered to the output locations in the shortest time possible, while respecting the ordering of different pallet types specified by the delivery list. The presented control system handles all aspects of warehouse operation, from individual vehicle control to high-level mission planning and coordination. Delivery mission assignments are optimized using dynamic programming and simulated annealing techniques. Mission executions are coordinated using graph search methods and a modified version of the Banker's algorithm, to ensure safe, collision and deadlock-free system operation. System performance is evaluated on a virtual warehouse model, using the high fidelity USARSim simulator.

In this paper, we present a control architecture for fully automated warehouse operation. The scenario we are considering consists of several autonomous guided vehicles (AGVs), transporting packages from an input location to one of several destinations, through an unstructured warehouse environment. The AGVs are equipped with wheel encoders, laser range finders and RFID tag readers. Each package has a RFID tag encoding its destination. A map of the warehouse is provided a priori. We have developed a layered control architecture that enables fully autonomous system operation. Starting with a map of the environment, the control system generates feasible paths, assigns tasks to the vehicles and schedules task execution. At the lowest layer, waypoint tracking controllers enable task execution. The proposed architecture has been successfully tested in a simulated scenario at the IEEE Virtual Manufacturing Automation Competition (VMAC), using the USARSim simulator.

In this article we present a mathematical model of a spincopter - an inherently stable and easy to control unmanned aerial vehicle (UAV). Although the devised model represents only the first approximation of spincopter's physics, it captures basic phenomena that are sufficient for derivation and testing of a simple control algorithm. Results obtained by simulations in Simulink®, presented herein, demonstrate efficient performance of the proposed simple controller structure. Further tests, including investigation of complex aerodynamics effects and hardware-specified limitations, have been done using X-Plane®software package. Finally, the proposed controller has been tested experimentally. Both, X-Plane®and experimental results confirmed the outcome from Simulink®.

In this paper, we introduce machine-job incidence (MJI) matrix that can be obtained from Steward sequencing matrix and Kusiak machine-part incidence matrix. Methods for determination of structural properties of free-choice multiple reentrant systems (FMRF) are proposed and an explanation on how the content (number of active jobs) of those structures can be controlled. This paper gives a new method on how to determine if allocation of resources, in a form of repeatable sequences, gives stable system behavior. Although efficiency of the proposed methods have been demonstrated on examples involving manufacturing workcells, the method can be used for other discrete-event systems as well, as long as the system under study belongs to FMRF class.

In this paper we present a formation synchronization and reconfiguration scheme based on distributed computation. In our previous work, we have introduced a formation abstraction by means of a virtual rectangular grid. The grid was used to ensure collision-free transitions between formations, coordinated by a centralized controller. This paper is an extension of that work, proposing control laws that are computed in a distributed way while still ensuring collision-free formation reconfigurations. Furthermore, we consider an extension of the group rendezvous problem. The agents are required to meet at the rendezvous point and establish consensus on formation state. Again, the grid abstraction is used to describe the formation state. Nonholonomic constraints of agent motion are taken into account explicitly. As our approach is communication-based, we also examine the effects of temporary communication loss.

This paper presents a dynamic routing method for supervisory control of multiple automated guided vehicles (AGVs) that are traveling within a layout of a given warehouse. In dynamic routing a calculated path particularly depends on the number of currently active AGVs' missions and their priorities. In order to solve the shortest path problem dynamically, the proposed routing method uses time windows in a vector form. For each mission requested by the supervisor, predefined candidate paths are checked if they are feasible. The feasibility of a particular path is evaluated by insertion of appropriate time windows and by performing the windows overlapping tests. The use of time windows makes the algorithm apt for other scheduling and routing problems. Presented simulation results demonstrate efficiency of the proposed dynamic routing. The proposed method has been successfully implemented in the industrial environment in a form of a multiple AGV control system.

In this paper we present a formation reconfiguration methodology designed for controlling groups of autonomous agents in environments densely populated with obstacles. Our approach is based on abstracting the group of agents by a discrete rectangular grid. Agent and obstacle positions are mapped onto the formation grid. Then, collision free formation transition trajectories are computed using discrete event scheduling techniques that have been well-established in the manufacturing systems domain. The main contribution of this paper is a unified formation control framework that explicitly takes obstacles into account. Using discrete event system analysis tools we show that our approach guarantees convergence to the desired formation while avoiding obstacles and inter-agent collisions.

The paper deals with an application of fuzzy control in joint servo control of a five-axis electric driven articulated robot arm. The same fuzzy control algorithm is used in all joint servo loops, requiring only multiplexing and proper scaling of the fuzzy controller inputs and outputs. The nonlinear character of the proposed fuzzy controller has provided for highly effective trajectory tracking. Trajectory tracking appears satisfactory also in the presence of external disturbances.>