Work

Editorial: Special Issue: 2nd Workshop on Integration of Vision and Inertial Sensors

Year: 2007
Type: editorial
Abstract: This special issue of The International Journal of Robotics Research is devoted to the topic of Integration of Visual and Inertial Sensors.For robots, as for animals, a sense of self-motion is critical to performing useful tasks in a complex and non-priory mapped world.The problems with sensing, particularly reliability and robustness, are well known to robotocists and sensor-fusion is a standard technique.Animals also make use of multiple sensory modalities, not just vision, smell, hearing, touch and taste but also joint position, muscle exertion, balance and motion.Some have even developed task-or environment-specific sensors such as acoustic ranging, magnetic dead-reckoning and electric fields.In this special issue we discuss two sensors that are common to animals and robots -vision and inertial sensing -which when combined provide rich information about self-motion and also the structure of the world.The advantage of these sensors is that they require no external infrastructure such as GPS satellites or visual or acoustic navigation beacons.Neither vision nor inertial sensors alone are perfect; however they have nice complementarities where each compensates for a shortcoming in the other.While inertial sensing is able to follow motions of any speed, its uncertainty -in particular at slow motions -is high compared to visual input.On the other hand visual sensors have problems with fast motions and find it difficult to distinguish between rotational and translational motion.Visual and inertial sensors today have high performance and are low cost and compact.They require no external reference and emit no radiation.Digital camera chips and micromachined accelerometers and gyroscopes are now commodities, and when combined with today's available computing can provide robust estimates of self-motion as well 3D scene structure.
Language: Spanish
Open Access status: bronze
OpenAlex ID: W2066520795
Cites:
Cited by: 27
Related to: 10