For autonomous mobile robots, a high precision positioning method in unstructured and dynamically changing environments is indispensable. Dead reckoning is a popular method but due to the error accumulation from wheel slippage, its reliability is low for measurement of long distances especially on uneven surfaces. Another popular method is the landmark method, which estimates current position relative to known landmarks, but the landmark method's limitation is that it cannot be used in an uncharted environment. GPS (Global Positioning System) has high performance as a three-dimensional positioning system on the ground level. However, it cannot be used in the environment where radio waves cannot reach such as indoors or underground. To overcome these shortcomings, we have been proposing a new method called "Cooperative Positioning System (CPS)". The main concept of CPS is to divide the robots into two groups, A and B where group A remains stationary and acts as a landmark while group B moves and then group B stops and acts as a landmark for group A. This process is repeated until the target position is reached. Compared with dead reckoning, CPS has a far lower accumulation of positioning errors, and can also work in three dimensions. Furthermore, CPS employs inherent landmarks and therefore can be used in uncharted environments unlike the landmark method. We have developed several CPS machine models named CPS-I, CPS-II and CPS-III, and demonstrated their performance for several tasks, such as floor cleaning and surveying in an unknown environment.
|CPS-I||CPS-II-||Floor cleaning experiment
in outdoor with CPS-IV
|Laser sensing robots, CPS-V|
|Laser sensing robots, CPS-V'||Laser sensing robots, CPS-VI||Laser sensing robots, CPS-VII|
We have been developing a robust localization technique for robocup middle league robots using omni-directional cameras.
We have been developing a rescue robot system consisting of multiple robots.
|Rescue robot system||Stable step climbing||3D model of train station|
|Motion control in 3D space||Motion control in 3D space||Step climbing using "tail"|