• DocumentCode
    2476838
  • Title

    Locating underground chemical sources by tracking chemical gradients in 3 dimensions

  • Author

    Russell, R.Andrew

  • Author_Institution
    Centre for Perceptive & Intelligent Machines in Complex Environ.: Intelligent Robotics, Monash Univ., Clayton, Vic., Australia
  • Volume
    1
  • fYear
    2004
  • fDate
    28 Sept.-2 Oct. 2004
  • Firstpage
    325
  • Abstract
    This paper describes current progress in a project to develop a robot for locating underground chemical sources. There are important economic, environmental and humanitarian applications for systems that can find the source of an odor or volatile chemical located beneath the ground. These applications cover the spectrum from harvesting truffles, to finding the location of leaking gas pipes and land mines. In previous work the feasibility of detecting vapor percolating through the ground was demonstrated. An algorithm was then developed to control a robot enabling it to gather information about the location of the source and to guide the robot towards it. This algorithm was limited to searching in two dimensions (depth was held constant). In this paper the original algorithm is extended to 3-dimensions allowing variations in depth to be included. The 3-dimensional search algorithm involves following a path through the ground that corresponds to the edges of closely packed dodecahedra. Simulations and physical experiments both confirm the capabilities of this search algorithm.
  • Keywords
    gas sensors; landmine detection; mobile robots; search problems; 3 dimensional search algorithm; chemical gradients tracking; land mine detection; leaking gas pipes; mobile robot; underground chemical sources location; Animals; Chemical hazards; Environmental economics; Intelligent robots; Leak detection; Machine intelligence; Robot sensing systems; Soil; Underwater tracking; Weapons;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Intelligent Robots and Systems, 2004. (IROS 2004). Proceedings. 2004 IEEE/RSJ International Conference on
  • Print_ISBN
    0-7803-8463-6
  • Type

    conf

  • DOI
    10.1109/IROS.2004.1389372
  • Filename
    1389372