M. Soldo. Reactive and pre-planned control in a mobile robot. In Proc. IEEE Int. Conf. Robotics and Automation, page 1128, May 1990.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....a closed loop control of mobile robots or vehicles has known several developments these last years. Two large classes of approaches can be distinguished : ffl the behavioral approach, consisting in directly relating inputs (or stimuli) to outputs (or actions) through systems state [Brooks, 1989, Soldo, 1990] The specicity of the input output linkage is however too restrictive for a general use. ffl the sensor based control approach, which directly introduces the sensory data in the robot closed loop control. The main benet consists in specifying the servoing problem directly in terms of a ....

Soldo, M. (1990). Reactive and preplanned control in a mobile robot. In Proc. of IEEE Int. Conf. on Robotics and Automation, pages 11281132.

....these systems for robotics systems control in real environments has been a substantial failure, due to the dynamic and uncertain nature of these domains. In these domains is difficult to completely synthesise a plan before its execution. To this end, since the late eighties many reactive planners [14, 10, 6] have been proposed. These systems use a pre compiled plan library instead of conIII structing a plan from primitive actions. As the planning phase is not distinct from the execution one, it s possible to dynamically increase the knowledge necessary to the plan complete construction. ....

M. Soldo. Reactive and preplanned control in a mobile robot. IEEE 1990.

....However, these systems assume that all of their behaviors are running concurrently on separate processors. In addition to the scaling problems discussed earlier, the processor per behavior scheme also wastes computing power, since many behaviors need not be active at various times. Other systems [8, 39] have made provisions to activate only subsets of behaviors, in much the same way as CIRCA runs different TAP schedules. However, these other systems do not reason about the resources required for each set of behaviors, and do not use advanced AI techniques to control the set of activated ....

M. H. Soldo, "Reactive and Preplanned Control in a Mobile Robot," in Proc. IEEE Int'l Conf. on Robotics and Automation, pp. 1128--1132, 1990.

.... we suggest a dichotomy of roles in the navigation system: a superior direction giver (mapmaker) and a dumb direction follower (navigator) Our qualitative environmental navigation system, which consists of the map maker and the navigator, can be categorized as both a preplanned and reactive type [65]. It is preplanned because the desired path of the navigator is precomputed by the map maker subsystem. But it is also reactive because the custom map provided by the map maker for the navigator does not hard code the desired path but rather describes what the navigator s sensors are supposed ....

M. H. Soldo. Reactive and preplanned control in a mobile robot. In Proceedings: Image Understanding Workshop, 1990.

No context found.

M. Soldo. Reactive and pre-planned control in a mobile robot. In Proc. IEEE Int. Conf. Robotics and Automation, page 1128, May 1990.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC