示例#1
0
    def kinematicCalculation(self, objMsg, joint):

        #	Get ball error
        ballError = objMsg.ball_error

        #	Get camera kinematics
        transformationMatrix = self.forwardKinematics(joint)

        #	If not find the ball
        if objMsg.ball_confidence == False:

            #	Set ball 3D Cartesion is None
            ball3DCartesian = None

        else:
            #	Calculate 3D coordinate respect to base frame
            ballPosition = np.array(objMsg.ball,
                                    dtype=np.float64).reshape(-1, 2)
            ball3DCartesian = self.calculate3DCoor(
                ballPosition, HCamera=transformationMatrix)
            ball3DCartesian = ball3DCartesian[0][1]

        #	If ball 3D cartesion is None
        if ball3DCartesian is not None:

            #	Calculate polar coordinate change x, y -> r, theta
            ball2DPolar = cv2.cartToPolar(ball3DCartesian[0],
                                          ball3DCartesian[1])
            ball2DPolar = np.array([ball2DPolar[0][0], ball2DPolar[0][0]])

        else:

            #	If can't calculate 3D return empty vector
            ball3DCartesian = np.array([])
            ball2DPolar = np.array([])

        rospy.logdebug(
            "\nPosition in 3D coordinate : {}\n".format(ball3DCartesian))

        #	Get 2D projection back to camera
        self.point2D1 = self.calculate2DCoor(self.points,
                                             "ground",
                                             HCamera=transformationMatrix)
        self.point2D1 = np.array(self.point2D1)

        #	Publist positon dict message
        msg = postDictMsg()
        msg.ball_cart = ball3DCartesian.reshape(-1)
        msg.ball_polar = ball2DPolar.reshape(-1)
        msg.ball_img = objMsg.ball
        msg.imgW = objMsg.imgW
        msg.imgH = objMsg.imgH
        msg.ball_error = ballError
        msg.ball_confidence = objMsg.ball_confidence
        msg.header.stamp = rospy.Time.now()

        return msg
示例#2
0
    def kinematicCalculation(self, objMsg, joint):

        #	Get ball error
        ballError = objMsg.ball_error

        #	Get camera kinematics
        # transformationMatrix = self.forwardKinematics( joint )
        qPan = getJsPosFromName(joint, "pan")
        qTilt = getJsPosFromName(joint, "tilt")
        transformationMatrix = getMatrixForForwardKinematic(qPan, qTilt)

        #	If not find the ball
        if objMsg.ball_confidence == False:

            #	Set ball 3D Cartesion is None
            ball3DCartesian = None

        else:
            #	Calculate 3D coordinate respect to base frame
            ballPosition = np.array(objMsg.ball,
                                    dtype=np.float64).reshape(-1, 2)
            ball3DCartesian = self.calculate3DCoor(
                ballPosition, HCamera=transformationMatrix)
            ball3DCartesian = ball3DCartesian[0][1]

        #	If ball 3D cartesion is None
        if ball3DCartesian is not None:

            #	Calculate polar coordinate change x, y -> r, theta
            ball2DPolar = cv2.cartToPolar(ball3DCartesian[0],
                                          ball3DCartesian[1])
            ball2DPolar = np.array([ball2DPolar[0][0], ball2DPolar[0][0]])

        else:

            #	If can't calculate 3D return empty vector
            ball3DCartesian = np.array([])
            ball2DPolar = np.array([])

        rospy.loginfo(
            "\nPosition in 3D coordinate : {}\n".format(ball3DCartesian))

        #	Publist positon dict message
        self.msg = postDictMsg()
        self.msg.ball_cart = ball3DCartesian.reshape(-1)
        self.msg.ball_polar = ball2DPolar.reshape(-1)
        self.msg.ball_img = objMsg.ball
        self.msg.imgW = objMsg.imgW
        self.msg.imgH = objMsg.imgH
        self.msg.ball_error = ballError
        self.msg.ball_confidence = objMsg.ball_confidence
        self.msg.header.stamp = rospy.Time.now()

        return self.msg
示例#3
0
    def kinematicCalculation(self, objMsg, js, cortexMsg, rconfig=None):
        t0 = time.time()

        pointsClound2D = []
        boundary2D = []
        object2D = []
        indexList = []

        for i, (p, n) in enumerate(zip(objMsg.pos2D, objMsg.object_name)):
            if n == 'point':
                pointsClound2D.append([p.x, p.y])
            elif n == 'field':
                boundary2D.append([p.x, p.y])
            else:
                object2D.append([p.x, p.y])
                indexList.append(i)
        t1 = time.time()

        pitch = math.radians(cortexMsg.pitch) if cortexMsg is not None else 0.0
        roll = math.radians(cortexMsg.roll) if cortexMsg is not None else 0.0
        # print math.degrees(pitch), math.degrees(roll)
        # roll = pitch = 0.0
        tranvec = np.zeros((3, 1))
        rotvec = np.array([roll, pitch, 0.0])

        HRotate = self.create_transformationMatrix(tranvec,
                                                   rotvec,
                                                   'rpy',
                                                   order="tran-first")

        H = getMatrixForForwardKinematic(js.position[0], js.position[1], roll,
                                         pitch)
        H = np.matmul(HRotate, H)

        t2 = time.time()
        pointsClound3D = self.calculate3DCoor(pointsClound2D, HCamera=H)
        t3 = time.time()
        boundary3D = self.calculate3DCoor(boundary2D, HCamera=H)
        t4 = time.time()
        object3D = self.calculate3DCoor(object2D, HCamera=H)
        t5 = time.time()

        polarList = []
        cartList = []
        names = []
        confidences = []
        errorList = []
        pos2DList = []

        landmarkPose3D = []
        ranges = []
        points = []
        splitIndexes = []

        imgH = objMsg.imgH
        imgW = objMsg.imgW

        for (plane, p3D), i in zip(object3D, indexList):
            name = objMsg.object_name[i]
            confidence = objMsg.object_confidence[i]
            p2D = objMsg.pos2D[i]
            err = objMsg.object_error[i]

            if name == 'magenta' or name == 'cyan':
                cartList.append(Point32(x=err.x, y=err.y))
                landmarkPose3D.append(Point32(x=err.x, y=err.y))

                polarList.append(Point32(x=err.x, y=err.y))

                names.append(name)
                confidences.append(confidence)
                errorList.append(Point32(x=err.x, y=err.y))
                pos2DList.append(p2D)

            elif plane is None:
                pass

            else:
                x, y = p3D[:2]

                cartList.append(Point32(x=x, y=y))
                landmarkPose3D.append(point2D(x=x, y=y))

                rho = math.sqrt(x**2 + y**2)
                phi = math.atan2(y, x)

                polarList.append(Point32(x=rho, y=phi))

                names.append(name)
                confidences.append(confidence)
                errorList.append(Point32(x=err.x, y=err.y))
                pos2DList.append(p2D)

        t6 = time.time()

        for plane, p3D in pointsClound3D:
            if plane is None:
                continue

            else:
                x, y = p3D[:2]
                # x *= 0.8
                # y *= 0.8
                ranges.append(math.sqrt(x**2 + y**2))
                points.append(point2D(x=x, y=y))

        splitIndexes.append(len(points))

        t7 = time.time()

        for plane, p3D in boundary3D:
            if plane is None:
                continue

            else:
                x, y = p3D[:2]
                # x *= 0.8
                # y *= 0.8
                ranges.append(math.sqrt(x**2 + y**2))
                points.append(point2D(x=x, y=y))

        circle = self.findCircle_ransac(np.array([[p.x, p.y] for p in points]))
        if circle is not None and circle[0] is not None and circle[
                1] is not None and circle[2] is not None:
            names.append('circle')
            landmarkPose3D.append(point2D(x=circle[0], y=circle[1]))
            confidences.append(
                self.getDensityProbability_normalize(circle[2], 0.75, 0.5))
        t8 = time.time()

        # print points3D
        msg = postDictMsg()
        msg.object_name = names
        msg.pos3D_cart = cartList
        msg.pos2D_polar = polarList
        msg.pos2D = pos2DList
        msg.object_error = errorList
        msg.object_confidence = confidences
        msg.imgH = imgH
        msg.imgW = imgW

        pointCloundMSG = scanlinePointClound()
        pointCloundMSG.num_scanline = 20
        pointCloundMSG.min_range = 0.5
        pointCloundMSG.max_range = 4
        pointCloundMSG.range = ranges
        pointCloundMSG.points = points
        pointCloundMSG.splitting_index = splitIndexes
        pointCloundMSG.jointState = js

        msgLocalize = localizationMsg()
        msgLocalize.pointClound = pointCloundMSG

        msgLocalize.landmark.names = names
        msgLocalize.landmark.pose = landmarkPose3D
        msgLocalize.landmark.confidences = confidences

        t9 = time.time()

        # print "Create npy :", t1 - t0
        # print "Creat H mat:", t2 - t1
        # print "Cal pointC :", t3 - t2
        # print "Cal boundy :", t4 - t3
        # print "Cal objMsg :", t5 - t4
        # print "Add obj2lis:", t6 - t5
        # print "Add poitntC:", t7 - t6
        # print "Add boundar:", t8 - t7
        # print "create msg :", t9 - t8
        # print "Overall	  :", t9 - t0
        # print '---------\n'

        return msg, msgLocalize
示例#4
0
    def kinematicCalculation(self, objMsg, js, cortexMsg, rconfig=None):

        self.points2D = [[p.x, p.y] for p in objMsg.pos2D]

        pitch = math.radians(cortexMsg.pitch) if cortexMsg is not None else 0.0
        roll = math.radians(cortexMsg.roll) if cortexMsg is not None else 0.0
        # print math.degrees(pitch), math.degrees(roll)
        # roll = pitch = 0.0
        tranvec = np.zeros((3, 1))
        rotvec = np.array([roll, pitch, 0.0])

        HRotate = self.create_transformationMatrix(tranvec,
                                                   rotvec,
                                                   'rpy',
                                                   order="tran-first")

        H = getMatrixForForwardKinematic(js.position[0], js.position[1], roll,
                                         pitch)
        H = np.matmul(HRotate, H)

        points3D = self.calculate3DCoor(self.points2D, HCamera=H)

        polarList = []
        cartList = []
        names = []
        confidences = []
        errorList = []
        pos2DList = []

        imgH = objMsg.imgH
        imgW = objMsg.imgW

        for (plane,
             p3D), name, confidence, p2D in zip(points3D, objMsg.object_name,
                                                objMsg.object_confidence,
                                                objMsg.pos2D):
            if plane is None:
                pass

            else:
                x, y = p3D[:2]

                if math.sqrt(x**2 + y**2) > 6:
                    continue

                cartList.append(Point32(x=x, y=y))

                rho = math.sqrt(x**2 + y**2)
                phi = math.atan2(y, x)

                polarList.append(Point32(x=rho, y=phi))

                names.append(name)
                confidences.append(confidence)
                errorList.append(Point32(x=p2D.x / imgW, y=p2D.y / imgH))
                pos2DList.append(p2D)
        # print points3D
        msg = postDictMsg()
        msg.object_name = names
        msg.pos3D_cart = cartList
        msg.pos2D_polar = polarList
        msg.pos2D = pos2DList
        msg.object_error = objMsg.object_error
        msg.object_confidence = confidences
        msg.imgH = imgH
        msg.imgW = imgW

        return msg
示例#5
0
	def kinematicCalculation( self, objMsg, joint ):
		
		#	get object name and passthrough
		objectNameList = objMsg.object_name

		#	Get camera kinematics
		# transformationMatrix = self.forwardKinematics( joint )
		qPan = getJsPosFromName( joint, "pan" )
		qTilt = getJsPosFromName( joint, "tilt" )
		transformationMatrix = getMatrixForForwardKinematic( qPan, qTilt )
		
		#	initial list
		ball3DCartesianList = list()
		ball2DPolarList = list()
		
		#
		#	convert for 'ball' only
		#
		
		#	loop every object to calculate 3D position
		for objIndex in range( len( objMsg.object_name ) ):
			
			if objMsg.object_confidence[ objIndex ] > 0.50:


				#	get 2D ball position
				ballPosition = np.array( [ objMsg.pos2D[ objIndex ].x, 
							   objMsg.pos2D[ objIndex ].y ], dtype = np.float64 )
				ballPosition = ballPosition.reshape( -1, 2 )


				#	get ball in world coordinate
				#	ball3DCartesian is returned in form [ ( 'projection', arrayOfPos ) ] 
				ball3DCartesian = self.calculate3DCoor( ballPosition, 
									HCamera = transformationMatrix )[ 0 ][ 1 ]
				#	get polar coordinate
				#	handle error when 3D coordinate is cannot calculate 3D position
				try:
					r = np.sqrt( ball3DCartesian[ 0 ] ** 2 + ball3DCartesian[ 1 ] ** 2 )
					theta = np.arctan2( ball3DCartesian[ 1 ], ball3DCartesian[ 0 ] )
					ball2DPolar = np.array( [ r, theta ] )

					#	compress to msg
					ball3DCartesianMsg = Point32( ball3DCartesian[ 0 ], 
					     			      ball3DCartesian[ 1 ], 
								      ball3DCartesian[ 2 ] )

					ball2DPolarMsg = Point32( ball2DPolar[ 0 ],
								  ball2DPolar[ 1 ],
						  				 0 )			
				except Exception as e:
					rospy.logwarn( e )
					ball3DCartesianMsg = Point32()
					ball2DPolarMsg = Point32()	
			else:

				ball3DCartesianMsg = Point32()
				ball2DPolarMsg = Point32()
				
			ball3DCartesianList.append( ball3DCartesianMsg )
			ball2DPolarList.append( ball2DPolarMsg )
			
		#	If not find the ball
#		if objMsg.ball_confidence == False or len( objMsg.ball ) == 0:
#			
#			#	Set ball 3D Cartesion is None
#			ball3DCartesian = None		
#
#		else:
#			#	Calculate 3D coordinate respect to base frame
#			ballPosition = np.array( objMsg.ball, dtype = np.float64 ).reshape( -1, 2 )
#			ball3DCartesian = self.calculate3DCoor( ballPosition, HCamera = transformationMatrix )
#			ball3DCartesian = ball3DCartesian[ 0 ][ 1 ]
#
#		#	If ball 3D cartesion is None
#		if ball3DCartesian is not None:
#
#			#	Calculate polar coordinate change x, y -> r, theta
#			#ball2DPolar = cv2.cartToPolar( ball3DCartesian[ 0 ], ball3DCartesian[ 1 ] )
#			#ball2DPolar = np.array( [ ball2DPolar[ 0 ][ 0 ], ball2DPolar[ 0 ][ 0 ] ] )
#			r = np.sqrt( ball3DCartesian[ 0 ] ** 2 + ball3DCartesian[ 1 ] ** 2 )
#			theta = np.arctan2( ball3DCartesian[ 1 ], ball3DCartesian[ 0 ] )
#			ball2DPolar = np.array( [ r, theta ] )
#
#		else:
#
#			#	If can't calculate 3D return empty vector
#			ball3DCartesian = np.array( [  ] )
#			ball2DPolar = np.array( [  ] )
#
#		rospy.loginfo( "\nPosition in 3D coordinate : {}\n".format( ball3DCartesian ) )
#		rospy.logdebug( "\nPosition in Polar coordinate : {}\n".format( ball2DPolar ) )
		
	 	#	Get 2D projection back to camera
		self.point2D1 = self.calculate2DCoor( self.points, "ground", HCamera= transformationMatrix )
		self.point2D1 = np.array( self.point2D1 )

 		#	Publist positon dict message
		msg = postDictMsg()
		msg.object_name = objectNameList
		msg.pos3D_cart = ball3DCartesianList
		msg.pos2D_polar = ball2DPolarList
		msg.pos2D = objMsg.pos2D
		msg.imgW = objMsg.imgW
		msg.imgH = objMsg.imgH
		msg.object_error = objMsg.object_error
		msg.object_confidence = objMsg.object_confidence
		msg.header.stamp = rospy.Time.now()

		return msg