random_mom_and_calf

pylint: disable=missing-docstring, global-statement, invalid-name, too-few-public-methods, no-self-use

A random mother cachelot and calf

Configuration that is common for the different cachelot examples.

from gzutils.gzutils import DotDict, get_output_dir

from sea import Squid

terrain = ('WWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWW\n' +
           'WWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWW\n' +
           'WWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWW\n' +
           'WWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWW\n' +
           'WWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWW\n' +
           'WWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWW\n' +
           'WWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWW')

the mother and calf have separate and identical lanes

things = ('XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX\n' +
          '                                                  \n' +
          '  ssss                                    ssss    \n' +
          'XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX\n' +
          '                                                  \n' +
          '  ssss                                    ssss    \n' +
          'XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX')

the mother and calf have separate and identical lanes

exogenous_things = ('                                                  \n' +
                    '                                                  \n' +
                    '  ssss                                    ssss    \n' +
                    '                                                  \n' +
                    '                                                  \n' +
                    '  ssss                                    ssss    \n' +
                    '                                                  ')


mom_start_pos = (0, 1)
calf_start_pos = (0, 4)

motors can perform several actions. The Sea Environment has four available actions: eat, down, up, forward. There is also one nsaction which is sing sensors are boolean variables indicating percepts (Things of different kinds) that are perceived. Active sensors are sent as input to the program

OPTIONS = DotDict({
    'output_path': get_output_dir(file=__file__),
    'terrain': terrain.split('\n'),
    'things': things.split('\n'),
    'exogenous_things': exogenous_things.split('\n'),
    'exogenous_things_prob': 0.005,
    'objectives': {'energy': 1.0},
    'rewards':{
        'sing_eat_and_forward': {
            Squid: {
                'energy': 0.1
            },
            None: {
                'energy': -0.05
            }
        },
        'eat_and_forward': {
            Squid: {
                'energy': 0.1
            },
            None: {
                'energy': -0.05
            }
        },
        'dive_and_forward': {
            None: {
                'energy': -0.002
            }
        },
        'up_and_forward': {
            None: {
                'energy': -0.001
            }
        },
        'forward': {
            None: {
                'energy': -0.001
            }
        },
        None: {
            None: {
                'energy': -0.001
            }
        },
    },
    'wss_cfg': {
        'numTilesPerSquare': (1, 1),
        'drawGrid': True,
        'randomTerrain': 0,
        'agents': {
            'mom': {
                'name': 'M',
                'pos': mom_start_pos,
                'hidden': False
            },
            'calf': {
                'name': 'c',
                'pos': calf_start_pos,
                'hidden': False
            }
        }
    }
})

random_mom_and_calf_config.py