Recent advances in technology are delivering robots of reduced size and cost. A natural outgrowth these systems comprised large numbers that collaborate autonomously diverse applications. Research on effective autonomous control such systems, commonly called swarms, has increased dramatically recent years received attention from many domains, as bioinspired robotics theory. These kinds distributed present novel challenges for the integration human supervisors, operators, teammates only...

We present distributed algorithms for multirobot task assignment where the tasks have to be completed within given deadlines. Each robot has a limited battery life and thus there is an upper limit on amount of time that it perform tasks. Performing each requires certain (called duration) can different payoffs Our problem assign robots such total payoff maximized while respecting deadline constraints robot's constraints. NP-hard since special case our classical generalized (which NP-hard)....

In this paper, we present provably-good distributed task assignment algorithms for a heterogeneous multi-robot system, in which the tasks form disjoint groups and there are constraints on number of robot can do (both within overall mission each group). Each obtains payoff (or incurs cost) objective allocation is to maximize (minimize) total (cost) robots. general, existing either assume that independent or not provide performance guarantee situation, exist. We algorithm an almost optimal...

Influence maximization in social networks has been widely studied motivated by applications like spread of ideas or innovations a network and viral marketing products. Current studies focus almost exclusively on unsigned containing only positive relationships (e.g. friend trust) between users. signed both negative foe distrust) users is still challenging problem that not studied. Thus, this paper, we propose the polarity-related influence (PRIM) which aims to find seed node set with maximum...

In this paper, we present an algorithm to solve the Multi-Robot Persistent Coverage Problem (MRPCP). Here, seek compute a schedule that will allow fleet of agents visit all targets given set while maximizing frequency visitation and maintaining sufficient fuel capacity by refueling at depots. We also heuristic method us bounded suboptimal results in real time. The produced our team robots efficiently cover or tasks persistently over long periods time, even when cost transition between is dynamic.

In this paper, we present provably-good algorithms for multi-robot task assignment, where each has to be completed within its deadline. Each robot a upper limit on the maximum number of tasks that it can perform due limited battery life, and takes same amount time complete. different payoff (or cost) objective is assign robots such total (cost) maximized (minimized) while respecting deadline constraints. This problem an extension special generalized assignment (where consumes resource must...

Controlling a swarm of robots after deployment is difficult, due to the unpredictable and emergent behavior algorithms. Past work has focused on influencing via statically selected leaders—swarm members that operator directly controls—that are pre-selected remain leaders throughout scenario execution. This paper investigates use dynamically controlled by human guide rest swarm, which operating under flocking-style algorithm. The goal move regions arise in environment. We experimentally...

Robotic swarms are distributed systems whose members interact via local control laws to achieve a variety of behaviors, such as flocking. In many practical applications, human operators may need change the current behavior swarm from goal that was going towards into new due dynamic changes in mission objectives. There two related but distinct capabilities needed supervise robotic swarm. The first is comprehension swarm's state and second prediction effects inputs on behavior. Both them very...

Robotic swarms are distributed systems that exhibit global behaviors arising from local interactions between individual robots. Each robot can be programmed with several control laws activated depending on an operator's choice of swarm behavior. While some simple (e.g. rendezvous) guaranteed performance known objectives under strict assumptions have been studied in the literature, real missions occur uncontrolled environments dynamically and require combinations behaviors. Given a library...

Abstract In this paper, we discuss the convergence of recent advances in deep neural networks (DNNs) with design robotic mechanisms, which entails conceptualization problem as a learning from space specifications to parameterization mechanisms. We identify three key inter-related problems that are at forefront using versatility DNNs solving mechanism problems. The first is representation mechanisms and their specifications, where challenges arise primarily non-Euclidean nature data. second...

In this paper, we present task allocation (assignment) algorithms for a multi-robot system where the tasks are divided into disjoint groups and there precedence constraints between groups. Existing auction-based assume independence hence can not be used directly to solve class of assignment problems that consider. our model, each robot do fixed number obtains benefit (or incurs cost) task. The only one from group. These arise when robots have set takes same time completed. We extend auction...

Distributed algorithms for (re)configuring mobile sensors to cover a given area are important autonomous multi-robot operations in application areas such as surveillance and environmental monitoring. Depending on the assumptions about choice of environment, sensor models, coverage metric, motion models nodes, there different versions problem that have been formulated studied. In this paper, we consider system holonomic robots equipped with anisotropic (e.g., limited field view cameras)...

A key challenge to create a sustainable and energy-efficient society is in making consumer demand adaptive energy supply, especially renewable supply. In this paper, we propose partially-centralized organization of consumers, namely, cooperative for purchasing electricity from the market. We novel multiagent coordination algorithm shape consumption cooperative. cooperative, central coordinator buys whole group consumers make their own decisions based on private constraints preferences. To...

We present a distributed on-line coordinated motion planning approach for group of mobile robots moving amidst dynamic obstacles. The objective the is to minimize total distance traveled by as well danger deadlock. Kinematic constraints, robot-obstacle collision avoidance and velocity/acceleration constraints are explicitly considered in individual robot's planner. A priority based scheme proposed deal with pair-wise inter-robot constraints. In particular, we model assignment into minimum...

We propose a geometric method to solve inverse kinematics (IK) problems of 7-DoF manipulators with joint offsets at shoulder, elbow, and wrist. Traditionally, position for redundant are solved by using an iterative based on the pseudo-inverse manipulator Jacobian. This provides single solution among infinitely many possible solutions IK problem manipulators. There no closed-form multiple offsets. Using our we can compute two-parameter search exploiting geometry structure manipulator. Our...

This article formulates gantry real-time scheduling in a work cell, where the material transfer is driven by gantries, as Markov decision process (MDP). Classical learning methods and planning for solving optimization problems MDP are discussed. An innovative method, called "Q-ADP," proposed to integrate reinforcement (RL) with approximate dynamic programming (ADP). Q-ADP uses model-free Q-learning algorithm learn state values through interactions environment, meanwhile, steps during opt ADP...

As swarms are used in increasingly more complex scenarios, further investigation is needed to determine how give human operators the best tools properly influence swarm after deployment. Previous research has focused on relaying from operator swarm, either by broadcasting commands entire or influencing through teleoperation of a leader. While these methods each have their different applications, there been lack into should be propagated leader-based methods. This paper focuses two simple...

We present a provably-good distributed algorithm for generalized task assignment problem in the context of multirobot systems, where robots cooperate to complete set given tasks. In multi-robot (MR-GAP), each robot has its own resource constraint (e.g., energy constraint), and needs consume certain amount obtain payoff task. The objective is find maximum tasks such that assigned at most one while respecting robots' constraints. MR-GAP NP-hard problem. It an extension linear since different...

In this paper, we develop a principled method to model line and surface contact with point contacts that is consistent physics-based models of (line) contact. We solve the detection dynamic simulation step simultaneously by formulating problem as mixed nonlinear complementarity problem. This allows us compute centre pressure well forces at (consistent friction model) along configuration velocities rigid objects. present geometrically implicit time-stepping scheme for between two bodies...

This article presents an analytical method to modulate the dynamic response of a robotic manipulator interacting with its environment when performing impedance-related tasks, through choice stiffness and damping parameters. By joint space analysis vibration experiments, we prove that in order preserve desired behavior robot Cartesian space, neither nor matrix can be arbitrarily chosen; this has meet criteria for any given configuration. After mapping parameters (matrices) into analyze...

Accurate dynamic simulation with robust handling of intermittent contact is necessary for a wide range robotics problems, including the design parts feeding devices, manipulation and kinodynamic planning, designing grasp strategies. In this paper we present an implicit time-stepping scheme multibody systems by incorporating constraints as set complementarity algebraic equations within dynamics model. We model each body intersection convex inequalities write between force distance function...