Multi-robot

Processing of task with splitting and parallel processing Multi-robot assembly M-M Mandatory, parallel... 

Multi-robot systems, e.g., a fleet of wheelchairs (Philips et al. 2011)... 

Rakowsky, U. K. Schneeweiss, W. G. 2004. An Approach to Reliability-Adaptive Multi-Robot Operation. ESA Communication Production Office Advanced Space Technologies for Robotics and Automation. Proceedings of the 8th ASTRA Workshop, Noordwijk/The Netherlands, WPP-236. Noordwijk ESA/ESTEC, pp. 299-306. 

Task allocation for multiple robots addresses the problem of assigning a set of tasks to the corresponding robots while achieving some specific objectives of team performance. Research in this field dates back to the late 1980s. Task allocation of multiple robots in dynamic environments is core to multi-robot control for a number of real-world applications, such as military [1], transport services [2], search and rescue [3], etc. 

The working environments of task allocation can be static or dynamic [4]. Static task allocation assumes completely known information about the environment, such as the number of tasks and robots, the arrival time of tasks, and the process of task execution. Traditionally, applications in multi-robot domains have largely remained in static scenarios, with an aim to minimise a cost function, such as total path length, or execution time of the team. Obviously, static approaches cannot adapt to changes in a dynamic environment. 

With reference to the taxonomy of Gerkey and Mataric [6], multi-robot task allocation problems can be classified along three dimensions. In the dimension of robot, it can be a single-task robot or a multi-task one. A single-task robot is capable... 

Multi-robot task allocation is a typically NP-hard problem. Its challenges become even more complicated when operations in uncertain environments, such as unexpected interference between robots, stochastic task requests, inconsistent information, and various component failures, are considered [4]. In such cases, it is not worth spending time and resources to secure an optimal solution, if the solutimi keeps changing as operations go on. Moreover, if time-window constraints are imposed, there may not be enough time to compute an exact and global solution. 

The basic objective of a multi-robot task allocation problem is to have tractable planning that produces efficient and practicable solutions. Auction-based, or market-based, approaches manage this by assembling team information at a single location to make decisions about assigning tasks over the team to produce practicable solutions quickly and concisely [8]. 

The earliest example of auction-based multi-robot coordination, called contract net protocol [11], appeared about 30 years ago. Auction-based multi-robot coordination approaches have been growing in popularity in recent years. They have been successfully implemented in a variety of domains, such as robotic transport [12, 13], mapping and exploration [14], house cleaning [15], and reconnaissance [16]. Auction-based approaches are preferable in online applications in that they... 

The closed-loop bid adjustment mechanism is incorporated with a multi-robot dynamic task allocation module in a simulator, which also includes a module for motion planning of a fleet of robots. The details of this motion planning approach can be found in [21]. 

Dias MB, Zlot R, Kalra N, Stentz A (2006) Market-based multi-robot coordination a survey and analysis. Proc IEEE 94(7) 1257-1270... 

Lerman K, Jones C, Galstyan A, Mataric MJ (2006) Analysis of dynamic task allocation in multi-robot systems. Int J Robot Res 25(3) 225-241... 

Mosteo AR, Montano L (2007) Comparative experiments on optimization criteria and algorithms for auction based multi-robot task allocation. In Proceedings of IEEE international conference on robotics and automation, Roma, Italy, pp 3345-3350... 

Jones EG, Dias MB, Stentz A (2011) Time-extended multi-robot coordination for domains with intra-path constraints. Auton Robot 30(4) 41-56... 

Khan MT, hnanuel T, de Silva CW (2010) Autonomous market-based multi-robot cooperation. hi Proceedings of international conference on intelligent and advanced systems, Malaysia, ppl-6... 

Gahrszka, A. Swiemiak, A. 2005. Non-cooperative game approach to multi-robot planning. International Journal of Applied Mathematics and Comuter Science 15(3) 359-367,... 

Gerkey, B.P., Vaughan, R.T., Howard, A. The Player/Stage Project Tools for Multi-Robot and Distributed Sensor Systems. In Proceedings of the Inti. Conf. on Advanced Robotics (ICAR), Coimbra, Portugal, pp. 317-323 (2003)... 

Multi-robot cooperative localization is a challenging task in mobile robotics, and it is the basis for both multi-robot navigation and exploration. Mobile platforms in the team are able to cooperate and estimate their poses in the environment. Robots in a cooperative localization system are able to make full use of their sensor information to refine their internal estimation and to improve their localization accuracy. Furthermore, the ability to exchange information during localization is particularly attractive, because each sight of another robot reduces the uncertainty of the estimated poses. This means that teams have to be designed to build a network infrastructure in order to perform cooperative tasks. 

Let us consider a set of mobile robots moving in a known environment. The multi-robot localization problem can be solved by discovering the pose (i.e., position andheading direction) ofeach single mobile platform. 

Throughout this chapter, the state space approach is chosen to model dynamic systems, together with a discreet-time formulation of the problem. In multi-robot localization, the state vector represents the poses of the robots comprising the team. In order to inspect inferences about a dynamie system, at least two models are required first, a model deseribing the time evolution of the state, i.e. the system dynamie model or state transition model, and seeond, a model deseribing the relation between die noisy measmements and the state, i.e. the measurement or observation model. 

To set up an effective multi-robot localization procedure, an online estimation is mandatory. In this case a recursive filter is a convenient solution, since received data can be processed sequentially rather than as a batch and there is no need... 

The multi-robot localization solution given by Equations (3) and (4) is only conceptual, since, in general, it cannot be resolved analytically. First of all, the implementation of Equations (3) and (4) requires the storage of non-Gaussian pdf, which is, in general terms, equivalentto an infinite... 

DISTRIBUTED MULTI-ROBOT LOCALIZATION USING lEKF... 

This chapter deals with multi-robot localization using a probabilistic approach. In this framework, each robot takes advantage of the integration of measurements retrieved by teammates in a collaborative scheme. However, maintaining a single posterior pdf over all robot poses leads to an unpractical solution, due to the high computational load. To overcome such drawbacks and distribute the estimate computation over the entire team, the posterior pdf can be factorized as ... 

A fully decentralized multi robot algorithm is designed exploiting lEKF. Each robot in the team runs its own localization procedure and exchanges pose information with the robots in the surrounding area. The state transition model for each sub-filter is given by ... 

This chapter deals with the design of a completely decentralized and distributed multi-robot localization algorithm. The localization procedure is based on nonlinear filtering. Specifically, it is an interlaced version of the EKF, able to produce accurate pose estimation for each robot of the team, thereby considerably reducing the computational load. 

---