def map_message(self):
        """ Return a nav_msgs/OccupancyGrid representation of this map. """

        grid_msg = OccupancyGrid()
        grid_msg.header.stamp = rospy.Time.now()
        grid_msg.header.frame_id = "map"

        grid_msg.info.resolution = self.occ_map.map_res
        grid_msg.info.width = self.occ_map.width
        grid_msg.info.height = self.occ_map.height

        grid_msg.info.origin = Pose(Point(self.occ_map.map_limit[0], self.occ_map.map_limit[2], 0),
                                    Quaternion(0, 0, 0, 1))

        flat_grid = copy.deepcopy(self.map.reshape((self.occ_map.map_size,)))
        for i in range(self.occ_map.map_size):
            if flat_grid[i] > self.occ_map.prob_occ:
                flat_grid[i] = 100
            elif flat_grid[i] < self.occ_map.prob_free:
                flat_grid[i] = 0
            else:
                flat_grid[i] = -1
        flat_grid = np.round(flat_grid)
        flat_grid = flat_grid.astype(int)
        grid_msg.data = list(flat_grid)
        return grid_msg

        def setUp(self):

            og = OccupancyGrid()
            og.info.resolution = 0.5
            og.info.height = 20
            og.info.width = 20
            og.info.origin.position.x = 0.0
            og.info.origin.position.y = 0.0
            og.info.origin.position.y = 0.0
            og.info.origin.orientation.z = 0.0
            og.info.origin.orientation.w = 0.0
            og.info.origin.orientation.x = 0.0
            og.info.origin.orientation.y = 0.0

            mapArray = np.array([0] * 400)
            mapArray.shape = (20, 20)

            for xi in range(20):
                for yi in range(20):
                    if xi < 3 or xi > 15:
                        mapArray[xi, yi] = 100
                    if yi < 3 or yi > 15:
                        mapArray[xi, yi] = 100
                    if yi < 10 and xi > 10:
                        mapArray[xi, yi] = 75

            mapArray.shape = 400
            og.data = list(mapArray)

            self.mapData = MapData(og)

def occ_grid_cb(data):
    global p,ready,first
    rospy.loginfo("map_reduction.py:occ_grid_cb() /ready_for_wps: %s",str(ready))
    if not ready.data:
	return
    nav_map = OccupancyGrid()
    w,h = data.info.width,data.info.height
    pt = data.data
    p = array(pt).reshape(w,h)#.T #<-- Transposes
    if not 'grid_loc' in globals():
        return
    x,y  = grid_loc.position.x, grid_loc.position.y
    nav_map.info = data.info
    if not 'size' in globals():
	size = 64
    if 'rtl' in globals() and rtl:
        ##Intelligently set mapsize, scale map, and send waypoints to home
	pass
    nav_map.info.width = size
    nav_map.info.height = size
    tmp_map = p[int(x)-size/2:int(x)+size/2,int(y)-size/2:int(y)+size/2]
    if first:
        tmp_map[:,:size/2] = 100
    nav_map.data = tmp_map.flatten()
    nav_map_pub.publish(nav_map)

    def updateCoverage(self):
        
        # Reinitialize coverage map
        ogm_shape = self.ogm.shape
        self.coverage = numpy.zeros(ogm_shape)
        
        # YOUR CODE HERE ------------------------------------------------------
        # Update the coverage field using the self.robot_path veriable.
        # We suppose that in every pose the robot covers an area of 2m x 2m
        # around it
        # 0 is for the uncovered, 100 is for the covered
        # PS. Try to make it fast :)
        # PS2. Do not have coverage values on obstacles or unknown space!
        # If done correctly, the coverage will appear purple in rviz

        # ---------------------------------------------------------------------
        # Publishing coverage ogm to see it in rviz
        coverage_ogm = OccupancyGrid()
        coverage_ogm.header.frame_id = "map"
        coverage_ogm.info = self.ogm_info
        coverage_ogm.data = numpy.zeros(self.ogm_info.width * self.ogm_info.height)
        for i in range(0, self.ogm_info.width):
            for j in range(0, self.ogm_info.height):
                coverage_ogm.data[i + self.ogm_info.width * j] = self.coverage[i][j]

        self.coverage_publisher.publish(coverage_ogm)

def processOccupancyGrid(gridMessage, scaleFactor, cacheEmptyCells):
    grid = OccupancyGrid(gridMessage.data, gridMessage.info.height, gridMessage.info.width, gridMessage.info.resolution,
                (gridMessage.info.origin.position.x, gridMessage.info.origin.position.y))
    grid.scale(scaleFactor, cacheEmptyCells=cacheEmptyCells)
    grid.expandObstacles()

    return grid

    def to_message(self):
        """ Return a nav_msgs/OccupancyGrid representation of this map. """
     
        grid_msg = OccupancyGrid()

        # Set up the header.
        grid_msg.header.stamp = rospy.Time.now()
        grid_msg.header.frame_id = "map"

        # .info is a nav_msgs/MapMetaData message. 
        grid_msg.info.resolution = self.resolution
        grid_msg.info.width = self.width
        grid_msg.info.height = self.height
        
        # Rotated maps are not supported... quaternion represents no
        # rotation. 
        grid_msg.info.origin = Pose(Point(self.origin_x, self.origin_y, 0),
                               Quaternion(0, 0, 0, 1))

        # Flatten the numpy array into a list of integers from 0-100.
        # This assumes that the grid entries are probalities in the
        # range 0-1. This code will need to be modified if the grid
        # entries are given a different interpretation (like
        # log-odds).
        flat_grid = self.grid.reshape((self.grid.size,)) * 100
        grid_msg.data = list(np.round(flat_grid))
        return grid_msg

def atlaas_grid(filepath, var_threshold=0.1, stamp=None, frame_id="/map"):
    """ Usage:
    atlaas.merge("atlaas.*x*.tif", "out.tif")
    if os.path.isfile("out.tif"):
        atlaas_grid("out.tif")
    """
    g = gdal2(filepath)
    bnp = g.bands[g.names["N_POINTS"]]
    bma = g.bands[g.names["Z_MAX"]]
    bme = g.bands[g.names["Z_MEAN"]]
    bva = g.bands[g.names["VARIANCE"]]
    data = numpy.where(bva > var_threshold, bma, bme)
    delta = data.max() - data.min()
    ddis = (100.0 * (data - data.min()) / delta).astype("uint8")
    ddis[bnp < 1] = 0  # no points, unknown
    og = OccupancyGrid()
    og.data = ddis.flatten()
    if stamp:
        og.header.stamp = stamp
    og.header.frame_id = frame_id
    og.info.resolution = g.scale_x
    og.info.height, og.info.width = data.shape
    og.info.origin.position.x, og.info.origin.position.y = g.u2c(g.utm_x, g.utm_y)
    og.info.origin.orientation.x = 1  # flip to transform UTM-ROS (scale_y < 0)
    return og

    def load_map(fp):
        
        ogrid = OccupancyGrid()
        step = None
        offset = None
        yaml_file = fp + ".yaml"
        csv_file = fp + ".csv"
        with open(yaml_file, "r") as infile:
            for k, v in yaml.load(infile.read()).iteritems():
                if k == "Header":
                    ogrid.header = v

                elif k == "MapMetaData":
                    ogrid.info = v
                    step = v.resolution
                    offset = (v.origin.position.x, v.origin.position.y)

                elif k == "GridFile":
                    print "GridFile: ", v
                else: 
                    print "Unexpected k : ", k
                    raise TypeError("Unexpected key type in yaml file: " + yaml_file)

            ogrid.data = list(np.loadtxt(csv_file, dtype=np.int8, delimiter=","))

        print "Map loaded"
        
        return ogrid, step, offset

def updateCombined():
	global data0
	global data1
	global data2
	global mapData
	global searchedCombinePub
	rospy.wait_for_message('robot_0/robotSearched', OccupancyGrid, timeout=None)
	rospy.wait_for_message('robot_1/robotSearched', OccupancyGrid, timeout=None)
	rospy.wait_for_message('robot_2/robotSearched', OccupancyGrid, timeout=None)
	mapMsg=OccupancyGrid()
	while not rospy.is_shutdown():
		map0 = numpy.array(data0.data)
		map1 = numpy.array(data1.data)
		map2 = numpy.array(data2.data)
		
		combined = numpy.minimum(map0,map1)
		combined = numpy.minimum(combined,map2)
	
		mapMsg.header.stamp=rospy.Time.now()
		mapMsg.header.frame_id=mapData.header.frame_id
		mapMsg.info.resolution=mapData.info.resolution
		mapMsg.info.width=mapData.info.width
		mapMsg.info.height=mapData.info.height
		mapMsg.info.origin=mapData.info.origin
		mapMsg.data=combined.tolist()
		
		searchedCombinePub.publish(mapMsg)
		
		rospy.sleep(1.0)

 def pubMap(self):
     # Publish updated map 
     grid_msg = OccupancyGrid()
     grid_msg.header.stamp = rospy.Time.now()
     grid_msg.header.frame_id = "map"
     
     # .info is a nav_msgs/MapMetaData message. 
     grid_msg.info.resolution = self.resolution
     grid_msg.info.width = self.map_info.width
     grid_msg.info.height = self.map_info.height
     
     # Rotated maps are not supported... quaternion represents no
     # rotation. 
     grid_msg.info.origin = Pose(Point(self.map_origin[0], self.map_origin[1], 0),Quaternion(0, 0, 0, 1))
     
     # Flatten the numpy array into a list of integers from 0-100.
     # This assumes that the grid entries are probalities in the
     # range 0-1. This code will need to be modified if the grid
     # entries are given a different interpretation (like
     # log-odds).
     flat_grid = self.grid.reshape((self.grid_map.size,)) * 100
     grid_msg.data = list(np.round(flat_grid))
     self._map_data_pub.publish(grid_msg.info)
     self._map_pub.publish(grid_msg)
     return grid_msg

def read_map(data):
    global workingOnMap
    workingOnMap = True
    print "manually reducing map resolution"
    # change the map's resolution
    newOccGrid, newMapHeight, newMapWidth = increaseMapResolution(data)
    # expand the obstacles 
    newOccGrid = expandObstacles(newOccGrid, desiredMapRes, newMapWidth, newMapHeight)

    # publish the new grid
    newPublishedGrid = OccupancyGrid()
    newPublishedGrid.info.resolution = desiredMapRes 
    newPublishedGrid.info.origin.position.x= data.info.origin.position.x
    newPublishedGrid.info.origin.position.y= data.info.origin.position.y
    newPublishedGrid.info.width = newMapWidth
    newPublishedGrid.info.height = newMapHeight
    newPublishedGrid.data = tuple(newOccGrid)
    print "publishing map"


    scaledMapPub.publish(newPublishedGrid)

    # publish the amount that the obstacles were expanded by
    expansionVal = Float64()
    expansionVal.data = expansionWidth

    obstacleExpansionDimensionPub.publish(expansionVal)

    workingOnMap = False

def handleSearchedCombinedMessage(data):
	global findFrontierPub
	#get map dimensions data
	
	width=data.info.width
	height=data.info.height
	searchedMap = data.data
	
	frontier = [0] * (width * height)
	
	for h in range(height):
		for w in range(width):
			if searchedMap[h*width+w] == -1:
				frontier[h*width+w]=-1
			if searchedMap[h*width+w] == 100:
				if isFrontier(searchedMap,h,w,height,width):
					frontier[h*width+w]=100
			
	mapMsg=OccupancyGrid()				
	mapMsg.header.stamp=rospy.Time.now()
	mapMsg.header.frame_id=data.header.frame_id
	mapMsg.info.resolution=data.info.resolution
	mapMsg.info.width=width
	mapMsg.info.height=height
	mapMsg.info.origin=data.info.origin
	mapMsg.data=frontier
	findFrontierPub.publish(mapMsg)

    def publish(self, publisher, real = False):
        costGrid = OccupancyGrid()

        costGrid.header.frame_id = self.frame_id
        costGrid.header.stamp = rospy.Time.now()
        costGrid.info = self.map_info
        temparr = copy.deepcopy(self.data) #protect ourselves from volitility

        if not real:
            # map cost_map to between 0 and 127 for fancy colors in rviz
            maxVal = max(temparr)

            minVal = float('inf')
            for cost in temparr:
                if not (cost == 0) and (cost < minVal): minVal = cost

            minVal *= 1.01

            factor = 90.0/(maxVal - minVal)

            if(maxVal == minVal): return

            # costGrid.data = [(int((i - minVal) * factor) if (i != 0) else 0) for i in cost_map.data]
            costGrid.data = []
            for i in temparr:
                if(i!=0):
                    costGrid.data.append(int((i - minVal) * factor))
                else:
                    costGrid.data.append(0)
        else: costGrid.data = temparr

        publisher.publish(costGrid)

def viz_data(grid, closed):
    # Visualize DATA
    global pub
    global grid_res, grid_transform
    occ_msg = OccupancyGrid()
    occ_msg.info.resolution = grid_res
    occ_msg.info.width = len(grid.nodelist)
    occ_msg.info.height = len(grid.nodelist[0])
    occ_msg.header.frame_id = "map"
    occ_msg.info.origin.position.x = grid_transform[0, 3]
    occ_msg.info.origin.position.y = grid_transform[1, 3]
    quaternion = tf.transformations.quaternion_from_matrix(grid_transform)
    occ_msg.info.origin.orientation.x = quaternion[0]
    occ_msg.info.origin.orientation.y = quaternion[1]
    occ_msg.info.origin.orientation.z = quaternion[2]
    occ_msg.info.origin.orientation.w = quaternion[3]
    occ_data = [0] * (occ_msg.info.width * occ_msg.info.height)
    width = occ_msg.info.width
    height = occ_msg.info.height
    # occ_data[width * start.x + start.y] = 100
    # occ_data[width * goal.x + goal.y] = 100
    top_cost = 1
    for node in closed:
        if node.g_cost > top_cost:
            top_cost = node.g_cost
    for node in closed:
        cost = node.g_cost / top_cost
        cost *= 225
        if cost > 99:
            cost -= 228
        occ_data[width * node.x + node.y] = cost
    occ_msg.data = occ_data
    pub.publish(occ_msg)

 def run(self):
     while self.is_looping():
         navigation_tf_msg = TFMessage()
         try:
             motion_to_robot = almath.Pose2D(self.motion.getRobotPosition(True))
             localization = self.navigation.getRobotPositionInMap()
             exploration_path = self.navigation.getExplorationPath()
         except Exception as e:
             navigation_tf_msg.transforms.append(self.get_navigation_tf(almath.Pose2D()))
             self.publishers["map_tf"].publish(navigation_tf_msg)
             self.rate.sleep()
             continue
         if len(localization) > 0 and len(localization[0]) == 3:
             navigation_to_robot = almath.Pose2D(localization[0][0], localization[0][1], localization[0][2])
             navigation_to_motion = navigation_to_robot * almath.pinv(motion_to_robot)
             navigation_tf_msg.transforms.append(self.get_navigation_tf(navigation_to_motion))
         self.publishers["map_tf"].publish(navigation_tf_msg)
         if len(localization) > 0:
             if self.publishers["uncertainty"].get_num_connections() > 0:
                 uncertainty = Marker()
                 uncertainty.header.frame_id = "/base_footprint"
                 uncertainty.header.stamp = rospy.Time.now()
                 uncertainty.type = Marker.CYLINDER
                 uncertainty.scale = Vector3(localization[1][0], localization[1][1], 0.2)
                 uncertainty.pose.position = Point(0, 0, 0)
                 uncertainty.pose.orientation = self.get_ros_quaternion(
                     almath.Quaternion_fromAngleAndAxisRotation(0, 0, 0, 1))
                 uncertainty.color = ColorRGBA(1, 0.5, 0.5, 0.5)
                 self.publishers["uncertainty"].publish(uncertainty)
         if self.publishers["map"].get_num_connections() > 0:
             aggregated_map = None
             try:
                 aggregated_map = self.navigation.getMetricalMap()
             except Exception as e:
                 print("error " + str(e))
             if aggregated_map is not None:
                 map_marker = OccupancyGrid()
                 map_marker.header.stamp = rospy.Time.now()
                 map_marker.header.frame_id = "/map"
                 map_marker.info.resolution = aggregated_map[0]
                 map_marker.info.width = aggregated_map[1]
                 map_marker.info.height = aggregated_map[2]
                 map_marker.info.origin.orientation = self.get_ros_quaternion(
                     almath.Quaternion_fromAngleAndAxisRotation(-1.57, 0, 0, 1))
                 map_marker.info.origin.position.x = aggregated_map[3][0]
                 map_marker.info.origin.position.y = aggregated_map[3][1]
                 map_marker.data = aggregated_map[4]
                 self.publishers["map"].publish(map_marker)
         if self.publishers["exploration_path"].get_num_connections() > 0:
             path = Path()
             path.header.stamp = rospy.Time.now()
             path.header.frame_id = "/map"
             for node in exploration_path:
                 current_node = PoseStamped()
                 current_node.pose.position.x = node[0]
                 current_node.pose.position.y = node[1]
                 path.poses.append(current_node)
             self.publishers["exploration_path"].publish(path)
         self.rate.sleep()

    def get_message(self):
        ogrid = OccupancyGrid()
        ogrid.header = navigator_tools.make_header(frame="enu")
        ogrid.info.resolution = self.resolution
        ogrid.info.height, ogrid.info.width = self.grid.shape
        ogrid.info.origin = self.origin
        grid = np.copy(self.grid)
        ogrid.data = np.clip(grid.flatten() - 1, -100, 100).astype(np.int8)

        return ogrid

    def pub_grid(self, *args):
        grid = self.grid_drawer.img

        ogrid = OccupancyGrid()
        ogrid.header.frame_id = '/enu'
        ogrid.header.stamp = rospy.Time.now()
        ogrid.info = self.map_meta_data
        ogrid.data = np.subtract(np.flipud(grid).flatten(), 1).astype(np.int8).tolist()

        self.ogrid_pub.publish(ogrid)

	def publishCombined(self):
		#Enter Main Loop
		while not rospy.is_shutdown():
			
			#Convert to Numpy  Arrays
			map = []
			for i in range(0, self.numRobots):
				map.append(numpy.array(self.searchedData[i].data))
				
			combined2 = map[0]
			if self.numRobots > 1:
				#Find Minimum of all maps
				for i in range(1, self.numRobots):
					combined2 = numpy.minimum(combined2,map[i])
					
			#Pack Occupancy Grid Message
			mapMsg=OccupancyGrid()
			mapMsg.header.stamp=rospy.Time.now()
			mapMsg.header.frame_id=self.mapData.header.frame_id
			mapMsg.info.resolution=self.mapData.info.resolution
			mapMsg.info.width=self.mapData.info.width
			mapMsg.info.height=self.mapData.info.height
			mapMsg.info.origin=self.mapData.info.origin
			mapMsg.data=combined2.tolist()
			
			#Convert combined Occupancy grid values to grayscal image values
			combined2[combined2 == -1] = 150			#Unknown -1->150 		(gray)
			combined2[combined2 == 100] = 255			#Not_Searched 100->255	(white)
														#Searched=0				(black)
														
			#Calculate percentage of open area searched
			numNotSearched = combined2[combined2==255].size
			numSearched = combined2[combined2==0].size
			percentSearched = 100*float(numSearched)/(numNotSearched+numSearched)
			percentSearchedMsg = Float32()
			percentSearchedMsg.data = percentSearched
			self.percentPub.publish(percentSearchedMsg)
			
			#Pack Image Message
			imageMsg=Image()
			imageMsg.header.stamp = rospy.Time.now()
			imageMsg.header.frame_id = self.mapData.header.frame_id
			imageMsg.height = self.mapData.info.height
			imageMsg.width = self.mapData.info.width
			imageMsg.encoding = 'mono8'
			imageMsg.is_bigendian = 0
			imageMsg.step = self.mapData.info.width
			imageMsg.data = combined2.tolist()
			
			#Publish Combined Occupancy Grid and Image
			self.searchedCombinePub.publish(mapMsg)
			self.imagePub.publish(imageMsg)
			
			#Update Every 0.5 seconds
			rospy.sleep(1.0)

 def to_message(self):
     """ Return a nav_msgs/OccupancyGrid representation of this map. """
     grid_msg = OccupancyGrid()
     grid_msg.header.stamp = rospy.Time.now()
     grid_msg.header.frame_id = "map"
     grid_msg.info.resolution = self.resolution
     grid_msg.info.width = self.width
     grid_msg.info.height = self.height
     grid_msg.info.origin = Pose(Point(self.origin_x, self.origin_y, 0), Quaternion(0, 0, 0, 1))
     grid_msg.data = self._numpy_to_data()
     return grid_msg

 def convert_to_occ_grid(self):
     res = OccupancyGrid()
     res.header.stamp = rospy.Time.now()
     res.header.frame_id = "odom_combined"
     res.info.resolution = 0.1
     res.info.width = 1000
     res.info.height = 1000
     res.info.origin.position.x = -50
     res.info.origin.position.y = -50
     res.data = self.grid
     return res