• Figure 9-12: 4-leg robot spider, turning skill
  • Figure 9-14: Wave generated for circular locomotion R Figure 9-13: Moving 6-Leg Robot, around the Circle
  • Ultra fast cnn based Hardware Computing Platform Concepts for adas visual Sensors and Evolutionary Mobile Robots




    Download 3,22 Mb.
    Pdf ko'rish
    bet68/81
    Sana16.05.2024
    Hajmi3,22 Mb.
    #238917
    1   ...   64   65   66   67   68   69   70   71   ...   81
    Bog'liq
    Alireza Fasih

    Figure 
    9-12: 4-leg robot spider, turning skill 
    01
    .
    4
    ,
    95
    .
    0
    02
    .
    3
    88
    .
    4
    49
    .
    2
    61
    .
    2
    2
    .
    3
    52
    .
    3
    05
    .
    3
    03
    .
    4
    ,
    35
    .
    2
    3
    .
    1
    23
    .
    2
    2
    .
    3
    68
    .
    1
    35
    .
    0
    64
    .
    1
    85
    .
    1
    73
    .
    1
    I
    T
    T
    B
    A


     
    111 
    According to Equation 9-7, the fitness value has a direct relation with the distance 
    between the initial position point (
    Init_Center_Pos
    ) and the robot position (
    Robot_Pos

    divided by (R). Further, this function has an inverse relation with the robot movement 
    (Robot_ Movement). 
    (9-7) 
    Fitness = (1 - Dist (Init_Center_Pos, Robot_Pos) / R) *(1/Robot_Movement)
     
    By optimizing the robot movement/locomotion, the fitness value will converge to zero. For 
    the case of an unstructured robot, we have designed a broken-leg spider. In this test, the 
    aim is learning the robot for the turning left and right skill as a complete and perfect 
    spider. In Figure 9-15 is shown the representation of this type of robot. 
    Figure 
    9-14: Wave generated for circular locomotion 

    Figure 
    9-13: Moving 6-Leg Robot, around the Circle 
    Path 
    Path 


     
    112 
    The definition of the fitness function is a little bit sensitive in this case. According to 
    Equation 9-8, the fitness value has a direct relation with the distance between the initial 
    position point (
    Init_Center_Pos
    ) and the robot position (
    Robot_Pos
    ). Further, this function 
    has an inverse relation with the robot angle (
    Robot_Angle
    ). During the optimization phase, 
    the aim/goal is converging the fitness function to zero. 
    (9-8)
    Fitness = Dist (Init_Center_Pos, Robot_Pos)*(1/Robot_Angle)

    (9-9) 
    After nearly 3300 chromosome generations and evolution, the robot is able to turn over 
    its Yaw axis. Figure 9-16 shows the result of the wave pattern for unstructured spider 
    turning. The usability of the templates in this paper can be summarized as follows. 
    Templates are stored in a list/memory. By a high level task management templates are 
    selected. This selection depends on the high level task management decision and the 
    environment situation as well. Further, a factor of high importance is the behavioral 
    architecture and behavioral programming. In fact, low level skills (e.g. Moving forward, 

    Download 3,22 Mb.
    1   ...   64   65   66   67   68   69   70   71   ...   81




    Download 3,22 Mb.
    Pdf ko'rish

    Bosh sahifa
    Aloqalar

        Bosh sahifa



    Ultra fast cnn based Hardware Computing Platform Concepts for adas visual Sensors and Evolutionary Mobile Robots

    Download 3,22 Mb.
    Pdf ko'rish