Miguel Tuazon edited Furthermore_thanks_to_the_arrival__.tex  about 8 years ago

Commit id: 85d3613e4d17c522a8514c42d155580466a5de87

deletions | additions      

       

The MRS is therefore in essence a technology demonstrator, using a plethora of interesting and atypical design and engineering solutions, such as the use of holonomic omni-directional (Meccanum) wheels, a combination of different sensors (tactile, ultrasound, infrared, optical, laser), redundant communication interfaces and others. As the main aim of this article is to discuss the collision-free manipulation of objects using a 3D optical system for localization and navigation purposes, we will introduce select technical and construction details of this robotic system.  As presented in MRS, in order to avoid problems or mistakes in overall performance of the robot,  the collision free manipulation as is  applied to the robot and it  has to be ensured assured  for the objects with rotation capabilities in order to avoid complications in overall performance of the robot. capabilities.  The collision detection is implemented on two levels, first is when the MRS researchers base it the collision detection  on the numerically determined determine  values using the inverse kinematic problem and second is when they use the measured values coming from the sensors. The Kinect camera sensors are used to find determine  the resulting distance information that is in the perpendicular position of the measured object from the face of theKinect  sensor. The output data are matrices measuring 640 × x  480 for both cameras. It is then necessary required  to extract the object of interest objects  from these matrices, and to calculate solve  itsposition  coordinates in its position,  relation to the face of the Kinect sensor. Based on the \citet{vachalek2015collision}'s  experiments andtheir  evaluation, the precision of the localization using the MS Windows Kinect optical system in the HSV color space for the MRS researchers' usage is satisfactory. usage.  Thus, considering the used low-cost technology indicates demonstrates  a fairly good precision, its use in mobile robotic robotics  can be well-justified.As presented in MRS, in order to avoid problems or mistakes in overall performance of the robot, the collision free manipulation is applied to the robot and it has to be assured for the objects with rotation capabilities. The collision detection is implemented on two levels, first is when the MRS researchers base the collision detection on the numerically determine values using the inverse kinematic problem and second is when they use the measured values coming from the sensors. The Kinect camera sensors are used to determine the resulting distance information that is in the perpendicular position of the measured object from the face of the sensor. The output data are matrices measuring 640 x 480 for both cameras. It is then required to extract the objects from these matrices, and and to solve its position