Sensor Architecture and Task Classification for Agricultural Vehicles and Environments 英文参考文献.docVIP

Sensors 2010, 10, 11226-11247; doi:10.3390/s101211226 OPEN ACCESS sensors ISSN 1424-8220 /journal/sensors Article Sensor Architecture and Task Classification for Agricultural Vehicles and Environments Francisco Rovira-Más Departamento de Ingeniería Rural y Agroalimentaria, Universidad Politécnica de Valencia, Camino de Vera s/n, 46022 Valencia, Spain; E-Mail: frovira@dmta.upv.es; Tel.: +34-963-877-291; Fax: +34-963-877-299 Received: 20 October 2010; in revised form: 26 November 2010 / Accepted: 1 December 2010 / Published: 8 December 2010 Abstract: The long time wish of endowing agricultural vehicles with an increasing degree of autonomy is becoming a reality thanks to two crucial facts: the broad diffusion of global positioning satellite systems and the inexorable progress of computers and electronics. Agricultural vehicles are currently the only self-propelled ground machines commonly integrating commercial automatic navigation systems. Farm equipment manufacturers and satellite-based navigation system providers, in a joint effort, have pushed this technology to unprecedented heights; yet there are many unresolved issues and an unlimited potential still to uncover. The complexity inherent to intelligent vehicles is rooted in the selection and coordination of the optimum sensors, the computer reasoning techniques to process the acquired data, and the resulting control strategies for automatic actuators. The advantageous design of the network of onboard sensors is necessary for the future deployment of advanced agricultural vehicles. This article analyzes a variety of typical environments and situations encountered in agricultural fields, and proposes a sensor architecture especially adapted to cope with them. The strategy proposed groups sensors into four specific subsystems: global localization, feedback control and vehicle pose, non-visual monitoring, and