Cooperative Object Transport in Multi-Robot Systems: A Review of the State-of-the-Art

Tuci, Elio; Alkilabi, Muhanad; Akanyeti, Otar

doi:10.3389/frobt.2018.00059

Cited by 177 publications

(74 citation statements)

References 92 publications

(130 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The problem of transporting objects with a combined system is also a typical cooperative object transport problem in multi-robot systems, which is a growing research interest in recent years. Tuci et al [22] provided a comprehensive summary for the scientific literature of cooperative object transport. Alonso-Mora et al [23] presented a method that exploits deformability during manipulation of soft objects by robot teams including three KUKA YouBot robots.…”

mentioning

confidence: 99%

Kinematic Modeling of a Combined System of Multiple Mecanum-Wheeled Robots with Velocity Compensation

Dai

et al. 2019

Sensors

View full text Add to dashboard Cite

In industry, combination configurations composed of multiple Mecanum-wheeled mobile robots are adopted to transport large-scale objects. In this paper, a kinematic model with velocity compensation of the combined mobile system is created, aimed to provide a theoretical kinematic basis for accurate motion control. Motion simulations of a single four-Mecanum-wheeled virtual robot prototype on RecurDyn and motion tests of a robot physical prototype are carried out, and the motions of a variety of combined mobile configurations are also simulated. Motion simulation and test results prove that the kinematic models of single-and multiple-robot combination systems are correct, and the inverse kinematic correction model with velocity compensation matrix is feasible. Through simulations or experiments, the velocity compensation coefficients of the robots can be measured and the velocity compensation matrix can be created. This modified inverse kinematic model can effectively reduce the errors of robot motion caused by wheel slippage and improve the motion accuracy of the mobile robot system.Sensors 2020, 20, 75 2 of 37 robot called MC-Drive, and the MC-Drive TP200 robot has been used to carry aircraft at Airbus manufacturing plants [13,14]. The KUKA omniMove UTV-2 set is a heavy-duty mobile platform with 12 Mecanum wheels that can be used to carry large objects [15]. (2) Using a combination of multiple Mecanum-wheeled robot platforms, which can be considered as a whole with cooperative omnidirectional motion and transport. Usually, mobile platforms used for cooperative transportation are symmetrically arranged. For example, a railcar body can be carried cooperatively by four KUKA omniMove mobile platforms at the Siemens plant in Krefeld, Germany [16]. The four mobile platforms are symmetrically arranged at four corners of the railcar body. Omni-directional mobile platforms with 8, 12, 16, or 32 Mecanum wheels can also be considered as specific combinations of multiple basic mobile platforms.The mature basic theory of four-Mecanum-wheeled robots is the basis of research on multi-robot systems. Muir [17][18][19] carried out basic research on Mecanum-wheeled robots and developed a kinematic and dynamic model and control on a four-Mecanum-wheeled robot. Campion et al. [20] studied structural properties and classification of kinematic and dynamic models of mobile wheeled robots and derived a motion constraint equation of a Mecanum wheel that can be used in kinematic research of multiple Mecanum-wheeled robots. Robots with four or more Mecanum wheels are overactuated systems with one or more motion constraints (wheel velocities are linearly correlated). Every additional wheel, and every additional robot, adds new motion constraints. So, the motion constraints of a multiple-Mecanum-wheeled robot system can be developed [21], which is also valid for multi-robot systems. The problem of transporting objects with a combined system is also a typical cooperative object transport problem in multi-robot systems, which is a growing rese...

show abstract

mentioning

confidence: 99%

Kinematic Modeling of a Combined System of Multiple Mecanum-Wheeled Robots with Velocity Compensation

Dai

et al. 2019

Sensors

View full text Add to dashboard Cite

show abstract

“…More researchers began to be interested in solving the problem of cooperative object transportation using MRS from the mid 90's [5], [13]- [16]. Although the settings could be varied a lot from each others, there are three major strategies to configure such a problem: 1) pushing-only strategy; 2) grasping strategy; 3) caging strategy [3]. Our research adopts the second strategy such that the robots coordinate with each others using different actions(pushing or pulling), while the robots are focusing on the transportation without considering the way they interact with the object.…”

Section: Related Workmentioning

confidence: 99%

“…Oversized object transportation is a typical task that is usually composed with smaller subtasks which can be assigned to multiple robots simultaneously [3]. We assume that the robots are physically attached to the object, and transport is achieved by either pushing or pulling (or both) the object.…”

Section: Introductionmentioning

confidence: 99%

Decentralized Control of Multi-Robot System in Cooperative Object Transportation Using Deep Reinforcement Learning

et al. 2020

View full text Add to dashboard Cite

Object transportation could be a challenging problem for a single robot due to the oversize and/or overweight issues. A multi-robot system can take the advantage of increased driving power and more flexible configuration to solve such a problem. However, an increased number of individuals also changed the dynamics of the system which makes control of a multi-robot system more complicated. Even worse, if the whole system is sitting on a centralized decision making unit, the data flow could be easily overloaded due to the upscaling of the system. In this research, we propose a decentralized control scheme on a multi-robot system with each individual equipped with a deep Q-network (DQN) controller to perform an oversized object transportation task. DQN is a deep reinforcement learning algorithm, thus does not require the knowledge of system dynamics, instead, it enables the robots to learn appropriate control strategies through trial-and-error style interactions within the task environment. Since analogous controllers are distributed on the individuals, the computational bottleneck is avoided systematically. We demonstrate such a system in a scenario of carrying an oversized rod through a doorway by a two-robot team. The presented multi-robot system learns abstract features of the task and cooperative behaviors are observed. The decentralized DQN-style controller is showing strong robustness against uncertainties. In addition, We propose a universal metric to assess the cooperation quantitatively.

show abstract

“…Several types of collective behaviours can be seen as instances of collective decision making (see Valentini et al, 2017Valentini et al, , 2016, including collective motion where robots have to agree on a common direction of motion, and aggregation where robots have to agree on a common location in the environment (see Garnier et al, 2005Garnier et al, , 2008Bayindir and Şahin, 2009;Correll and Martinoli, 2011;Gauci et al, 2014). Indeed, aggregation is often a necessity for many swarm robotic systems as it is a prerequisite for other cooperative behaviours (Dorigo et al, 2004;Tuci et al, 2018).…”

Section: Introductionmentioning

confidence: 99%

On self-organised aggregation dynamics in swarms of robots with informed robots

Firat

Ferrante

Gillet

et al. 2020

Neural Comput & Applic

Self Cite

View full text Add to dashboard Cite

Aggregation is a process observed in natural systems whereby individuals gather together to form large cluster. Recent studies with cockroaches and robots have shown that relatively simple individual mechanisms can account for how individuals manage to gather on a single shelter when two or more are available in the environment. In this paper, we use simulated swarms of robots to further explore the aggregation dynamics generated by these simple individual mechanisms. Our objective is to study the introduction of "informed robots", and to study how many of these are needed to direct the aggregation process toward a pre-defined site among those available in the environment. Informed robots are members of a group that selectively avoid the site/s where no aggregate should emerge, and stop only on the experimenter predefined site/s for aggregation. We study the aggregation process with informed robots in three different scenarios: two that are morphologically symmetric, whereby the different types of aggregation site are equally represented in the environment; and an asymmetric scenario, whereby the target site has an area that is half the area of the sites that should be avoided. We first show what happens when no robot in the swarm is informed: in symmetric environments, the swarm is able to break the symmetry and aggregates on one of the two 1

show abstract

Cooperative Object Transport in Multi-Robot Systems: A Review of the State-of-the-Art

Cited by 177 publications

References 92 publications

Kinematic Modeling of a Combined System of Multiple Mecanum-Wheeled Robots with Velocity Compensation

Kinematic Modeling of a Combined System of Multiple Mecanum-Wheeled Robots with Velocity Compensation

Decentralized Control of Multi-Robot System in Cooperative Object Transportation Using Deep Reinforcement Learning

On self-organised aggregation dynamics in swarms of robots with informed robots

Contact Info

Product

Resources

About