While many multi-robot systems rely on fortuitous cooperation between agents, some tasks, such as the assembly of large structures, require tighter coordination. We present a general software architecture for coordinating heterogeneous robots that allows for both autonomy of the individual agents as well as explicit coordination. This paper presents recent results with three robots with very different configurations. Working as a team, these robots are able to perform a high-precision docking task that none could achieve individually.

No abstract

Abstract

No abstract

In the process of automating an earthmoving machine, we have developed a model of soil-tool IntroductionAutomation of earthmoving offers improved efficiency, consistency in work quality and improved safety. This is particularly the case in mining sites where mass excavation is performed. Even reducing the execution time of a single cycle of an operation by a few seconds can translate into a large savings over the entire job. We have developed an autonomous excavator that is able to sense its environment, make plans, and execute trajectories to dig and to load trucks at speeds comparable to a human operator [14].Automation of earthmoving with an excavator, presents unique challenges. The dynamics of the mechanism and linkages are complex, there is uncertainty in the shape of the terrain and soil parameters, and, the interaction forces between the excavator and the environment are very large. Planning earthmoving operations requires the ability to model the effect of actions when the actions are carried out open-loop, and, the result of initial conditions when a closed loop controller is involved. A key problem in the modeling of digging trajectories is the estimation of resistive forces that are not only a function of the shape of the terrain, soil and tool parameters, but, also the controller that servos the joints using force feedback, to fill the bucket.In this paper we present a model of the interaction between the tool (the excavator bucket) and the terrain. This model reformulates a classical formula that has been proposed to model flat blades moving through flat ground, as in the case of a bulldozer blade or a tilling device used in flat terrain. Our model accounts for some phenomena that are particular to excavation. Since such models are not analytically invertible, we have developed a numerical scheme that uses measured data from excavation experiments to extract parameters of the model. These parameters are then used to predict resistive forces and bucket trajectories for future candidate actions [13]. In addition, since the soil parameters can vary significantly-the difficulty in digging often varies with the strata being excavated-it is necessary to determine the soilparameters on-line. We present results that shows the utility of a method that uses data from a small number of digs to predict forces for candidate digs in the future. Related WorkThere has been some research on the operation of earthmoving machinery [16][1] that explicitly addresses the issue of estimating forces necessary to overcome the shear strength of soil. Unfortunately, this work is mostly stated in empirical terms for specific types of machines and it is not clear how to extrapolate the methodology for arbitrary mechanisms. A considerable amount of research has explored finite element analysis of the soil plasticity, for example [2]. While this work performs a painstaking analysis of soil displacement, it is not suited for our purposes since it requires a very large number of iterations of numerical integration for e...