Abstract: To accelerate the remanufacturing process of electronic waste products, this study designs a partial disassembly line with the multi-robotic station to effectively dispose of excessive wastes. The multi-robotic partial disassembly line is a technical upgrade to the existing manual disassembly line. Balancing optimization can make the disassembly line smoother and more efficient. For partial disassembly line balancing with the multi-robotic station (PDLBMRS), a mixed-integer programming model (MIPM) considering the robotic efficiency differences is established to minimize cycle time, energy consumption and hazard index and to calculate their optimal global values. Besides, an enhanced NSGA-II algorithm (HNSGA-II) is proposed to optimize PDLBMRS efficiently. Finally, MIPM and HNSGA-II are applied to an actual mixed disassembly case of two types of computers, the comparison of the results solved by GUROBI and HNSGA-II verifies the correctness of the model and excellent performance of the algorithm, and the obtained Pareto solution set provides multiple options for decision-makers.
Abstract: In this paper, we describe how to achieve knowledge understanding and prediction (Situation Awareness (SA)) for multiple-agents conducting searching activity using Bayesian inferential reasoning and learning. Bayesian Belief Network was used to monitor agents' knowledge about their environment, and cases are recorded for the network training using expectation-maximisation or gradient descent algorithm. The well trained network will be used for decision making and environmental situation prediction. Forest fire searching by multiple UAVs was the use case. UAVs are tasked to explore a forest and find a fire for urgent actions by the fire wardens. The paper focused on two problems: (i) effective agents’ path planning strategy and (ii) knowledge understanding and prediction (SA). The path planning problem by inspiring animal mode of foraging using Lévy distribution augmented with Bayesian reasoning was fully described in this paper. Results proof that the Lévy flight strategy performs better than the previous fixed-pattern (e.g., parallel sweeps) approaches in terms of energy and time utilisation. We also introduced a waypoint assessment strategy called k-previous waypoints assessment. It improves the performance of the ordinary levy flight by saving agent’s resources and mission time through redundant search avoidance. The agents (UAVs) are to report their mission knowledge at the central server for interpretation and prediction purposes. Bayesian reasoning and learning were used for the SA and results proof effectiveness in different environments scenario in terms of prediction and effective knowledge representation. The prediction accuracy was measured using learning error rate, logarithm loss, and Brier score and the result proves that little agents mission that can be used for prediction within the same or different environment. Finally, we described a situation-based knowledge visualization and prediction technique for heterogeneous multi-UAV mission. While this paper proves linkage of Bayesian reasoning and learning with SA and effective searching strategy, future works is focusing on simplifying the architecture.
Abstract: We describe issues bedeviling the coordination of heterogeneous (different sensors carrying agents) multi-agent missions such as belief conflict, situation reasoning, etc. We applied Bayesian and agents' presumptions inferential reasoning to solve the outlined issues with the heterogeneous multi-agent belief variation and situational-base reasoning. Bayesian Belief Network (BBN) was used in modeling the agents' belief conflict due to sensor variations. Simulation experiments were designed, and cases from agents’ missions were used in training the BBN using gradient descent and expectation-maximization algorithms. The output network is a well-trained BBN for making inferences for both agents and human experts. We claim that the Bayesian learning algorithm prediction capacity improves by the number of training data and argue that it enhances multi-agents robustness and solve agents’ sensor conflicts.
Abstract: One of the main difficulties in developing multi-robot
systems (MRS) is related to the simulation and testing tools available.
Indeed, if the differences between simulations and real robots are
too significant, the transition from the simulation to the robot
won’t be possible without another long development phase and
won’t permit to validate the simulation. Moreover, the testing of
different algorithmic solutions or modifications of robots requires
a strong knowledge of current tools and a significant development
time. Therefore, the availability of tools for MRS, mainly with
flying drones, is crucial to enable the industrial emergence of these
systems. This research aims to present the most commonly used
tools for MRS simulations and their main shortcomings and presents
complementary tools to improve the productivity of designers in the
development of multi-vehicle solutions focused on a fast learning
curve and rapid transition from simulations to real usage. The
proposed contributions are based on existing open source tools as
Gazebo simulator combined with ROS (Robot Operating System) and
the open-source multi-platform autopilot ArduPilot to bring them to
a broad audience.
Abstract: This paper presents a new approach to control robots, which can quickly find their swarm while tracking a moving target through the obstacles of the environment. In this approach, an artificial potential field is generated between each free-robot and the virtual attractive point of the swarm. This artificial potential field will lead free-robots to their swarm. The swarm-finding of these free-robots dose not influence the general motion of their swarm and nor other robots. When one singular robot approaches the swarm then its swarm-search will finish, and it will further participate with its swarm to reach the position of the target. The connections between member-robots with their neighbors are controlled by the artificial attractive/repulsive force field between them to avoid collisions and keep the constant distances between them in ordered formation. The effectiveness of the proposed approach has been verified in simulations.
Abstract: Swarm principles are increasingly being used to design controllers for the coordination of multi-robot systems or, in general, multi-agent systems. This paper proposes a two-dimensional Lagrangian swarm model that enables the planar agents, modeled as point masses, to swarm whilst effectively avoiding each other and obstacles in the environment. A novel method, based on an extended Lyapunov approach, is used to construct the model. Importantly, the Lyapunov method ensures a form of practical stability that guarantees an emergent behavior, namely, a cohesive and wellspaced swarm with a constant arrangement of individuals about the swarm centroid. Computer simulations illustrate this basic feature of collective behavior. As an application, we show how multiple planar mobile unicycle-like robots swarm to eventually form patterns in which their velocities and orientations stabilize.
Abstract: A novel method using bearing-only SLAM to estimate node positions of a localization network is proposed. A group of simple robots are used to estimate the position of each node. Each node has a unique ID, which it can communicate to a robot close by. Initially the node IDs and positions are unknown. A case example using RFID technology in the localization network is introduced.
Abstract: In recent years a number of applications with multirobot
systems (MRS) is growing in various areas. But their design
is in practice often difficult and algorithms are proposed for the
theoretical background and do not consider errors and noise in real
conditions, so they are not usable in real environment. These errors
are visible also in task of target localization enough, when robots
try to find and estimate the position of the target by the sensors.
Localization of target is possible also with one robot but as it was
examined target finding and localization with group of mobile robots
can estimate the target position more accurately and faster. The
accuracy of target position estimation is made by cooperation of
MRS and particle filtering. Advantage of usage the MRS with particle
filtering was tested on task of fixed target localization by group of
mobile robots.
Abstract: In this paper, a new method of information fusion – DSmT (Dezert and Smarandache Theory) is introduced to apply to managing and dealing with the uncertain information from robot map building. Here we build grid map form sonar sensors and laser range finder (LRF). The uncertainty mainly comes from sonar sensors and LRF. Aiming to the uncertainty in static environment, we propose Classic DSm (DSmC) model for sonar sensors and laser range finder, and construct the general basic belief assignment function (gbbaf) respectively. Generally speaking, the evidence sources are unreliable in physical system, so we must consider the discounting theory before we apply DSmT. At last, Pioneer II mobile robot serves as a simulation experimental platform. We build 3D grid map of belief layout, then mainly compare the effect of building map using DSmT and DST. Through this simulation experiment, it proves that DSmT is very successful and valid, especially in dealing with highly conflicting information. In short, this study not only finds a new method for building map under static environment, but also supplies with a theory foundation for us to further apply Hybrid DSmT (DSmH) to dynamic unknown environment and multi-robots- building map together.
Abstract: In this paper, a Cooperative Multi-robot for Carrying
Targets (CMCT) algorithm is proposed. The multi-robot team
consists of three robots, one is a supervisor and the others are
workers for carrying boxes in a store of 100×100 m2. Each robot has
a self recharging mechanism. The CMCT minimizes robot-s worked
time for carrying many boxes during day by working in parallel. That
is, the supervisor detects the required variables in the same time
another robots work with previous variables. It works with
straightforward mechanical models by using simple cosine laws. It
detects the robot-s shortest path for reaching the target position
avoiding obstacles by using a proposed CMCT path planning
(CMCT-PP) algorithm. It prevents the collision between robots
during moving. The robots interact in an ad hoc wireless network.
Simulation results show that the proposed system that consists of
CMCT algorithm and its accomplished CMCT-PP algorithm
achieves a high improvement in time and distance while performing
the required tasks over the already existed algorithms.