Special Issue on Spatial Information Sensing
Masanori Idesawa, Editor
Senior ResearcherInformation Science Laboratory, RIKEN(The Institute of Physical and Chemical Research)2-1, Hirosawa, Wake, Saitama 351, Japan
Humans are able to respond flexibly to changes in the environment — grasping information of objects, environmental conditions and personal state — via senses. As such, it is indispensable for an automatic machine such as a robot to be equipped with a sensing system corresponding to the human senses which permits flexible work in a 3-D environment. Acquisition of 3-dimensional environmental information is important in particular, and spatial information measures are regarded as indispensable for acquiring 3-D environmental information. The visual sense plays an important role for humans to acquire spatial information. Reflecting on this, R&D on visual sensors is underway vigorously in the field of robotics, and many expectations are focusing on an optical sensing system which permits the detection of spatial information from remote positions without any influences on the object. Meanwhile, humans do not plan behavior based fully on sensor information, but have papered fixed form behavior programs which are launched with information obtained from sensors or implemented with some modification in many cases. There are several modes of sensing in human: a monitoring mode which detects generation of extraordinal state, and which does not always concentrate attention on all sensors to acquire information; an attentive sensing mode in which men concentrate attention on related sensors when an extraordinary phenomena is detected; intentional mode in which men detect intentionally information necessary for work or behavior to be implemented, or detect, intentionally and as planned, information being short in behavior programs; and so on. It is effective for robots when possible to acquire information in the same mode as humans and to control it the same as human behavior. Sensor data integration and sensor fusion are also required for more accurate judgement, in which much information is integrated, in addition to knowledge already accumulated; when each piece of information in itself is not enough for accurate judgement, integrated information may permit better judgement. Further, model base sensing which refers to an environmental model and knowledge based sensing, and which refers to accumulated knowledge will be important research subjects. These will be followed by intentional sensing which intentionally acquires information necessary for achievement of objectives. Finally, introduction of the hierarchical behavior mode, which corresponds to the human monitoring and attentive modes, is indispensable for a robot. In this mode, a sensor system detects the object and extraordinary state in monitoring mode, then analyses and processes details of phenomena of particularly interest in attentive mode, and then detects and acquires information necessary for achievement of objectives with optimal resolution in intentional sensing mode. The special issue on spatial information sensing is a compilation of articles on some spatial information acquisition methods, visual sensors and positioning sensors for robots. Also, some proposals are presented for realizing multiple resolution sensing, sensor fusion and active sensing, centering on an optical spatial information acquisition method.
This article is published under a Creative Commons Attribution-NoDerivatives 4.0 International License.