Robotics and Autonomous Systems xxx (2005) xxx–xxx Object recognition: A new application for smelling robots Amy Loutfi ∗ , Mathias Broxvall, Silvia Coradeschi, Lars Karlsson Center for Applied Autonomous Sensor Systems (AASS), ¨ Orebro University, 701-82 ¨ Orebro, Sweden Received 26 April 2004; received in revised form 2 June 2005; accepted 8 June 2005 Abstract Olfaction is a challenging new sensing modality for intelligent systems. With the emergence of electronic noses, it is now possible to detect and recognize a range of different odours for a variety of applications. In this work, we introduce a new application where electronic olfaction is used in cooperation with other types of sensors on a mobile robot in order to acquire the odour property of objects. We examine the problem of deciding when, how and where the electronic nose (e-nose) should be activated by planning for active perception and we consider the problem of integrating the information provided by the e-nose with both prior information and information from other sensors (e.g., vision). Experiments performed on a mobile robot equipped with an e-nose are presented. © 2005 Elsevier B.V. All rights reserved. Keywords: Electronic olfaction; Planning for perceptional actions; Anchoring; Sensing with multiple modalities 1. Introduction Electronic noses (e-noses) are now being used for a variety of research areas ranging from the quality evalu- ation in the foodstuff industry [30,17,37,5] to medical applications [13,22]. In these contexts, an electronic nose is defined as an array of chemical gas sensors and a pattern recognition component that discriminates and identifies both simple and complex odours [14]. Today, e-noses are commercially available [10,1], and there are reliable pattern recognition algorithms that ∗ Corresponding author. Fax: +46 19 30 34 63. E-mail address: amy.loutfi@tech.oru.se (A. Loutfi). URL: http://www.aass.oru.se. can be applied to the sensor data [26,24,15,3]. Given this development, it is possible to integrate the full capabilities of electronic noses on mobile robots. Currently in mobile robotics, research applications using gas sensing focus on trail following [34], local- ization of odour plume origin [18,12] and mapping of odour distributions [25,38]. While this is an interesting area of research, it has yet to use the full ability of elec- tronic olfaction, since often only one specific substance is used for which the gas sensor is especially sensi- tive. Having mobile robots that can discriminate among odours would enhance their perceptual abilities and make it possible to identify different substances, assess odour characteristics and recognize objects using the odour property. 0921-8890/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.robot.2005.06.002 ROBOT-1244; No. of Pages 18