Example #1
0
    def showgraspfrompickle(self, object_name,metric_name,onlynagative,onlypositive,openravechecker):    
        object = self.dataset[object_name] 
        labelgrasps,labelmetrics=self.readdatafrompickle(object_name)

        vis.figure()
        vis.mesh(object.mesh.trimesh ,style='surface')
        config=self._get_config(None)
        low =0.0  
        high = np.max([labelmetrics[i][metric_name] for i in range(len(labelmetrics))])
     #   print 'high',high,config['quality_scale']
        if low == high:
            q_to_c = lambda quality: config['quality_scale']
        else:
            
            q_to_c = lambda quality: config['quality_scale'] * (quality - low) / (high - low)

        recalculate_grasp=[]    
        for i in range(len(labelgrasps)):
                 
                        if labelmetrics[i][metric_name]==0 :
                           if not onlypositive:
                               vis.graspwithapproachvectorusingcenter_point(labelgrasps[i] ,approaching_color=(1,0,0), grasp_axis_color=(1,0,0) )
                        elif  labelmetrics[i][metric_name]==-1   :
                            if not onlypositive:
                                 vis.shownormals(labelgrasps[i][1],labelgrasps[i][0],color=(1,0,0) )
                                 recalculate_grasp.append(labelgrasps[i])
                        elif  labelmetrics[i][metric_name]>0   :
                                if not onlynagative:
                                    color = plt.get_cmap('hsv')(q_to_c(labelmetrics[i][metric_name]))[:-1] 
                                    vis.graspwithapproachvectorusingcenter_point(labelgrasps[i] ,approaching_color=color, grasp_axis_color=color )
                                recalculate_grasp.append(labelgrasps[i])
                        print i   # ,high
                                
        vis.pose(RigidTransform(), alpha=0.1)
        vis.show(animate=False)       
Example #2
0
    def compute_topple(self, state):
        # raise NotImplementedError
        mesh = state.mesh.copy().apply_transform(state.T_obj_world.matrix)

        mesh.fix_normals()
        vertices, face_ind = sample.sample_surface_even(mesh, self.num_samples)
        # Cut out vertices that are too close to the ground
        z_comp = vertices[:,2]
        valid_vertex_ind = z_comp > (1-self.thresh)*np.min(z_comp) + self.thresh*np.max(z_comp)
        vertices, face_ind = vertices[valid_vertex_ind], face_ind[valid_vertex_ind]

        normals = mesh.face_normals[face_ind]
        push_directions = -deepcopy(normals)

        self.toppling_model.load_object(state)
        poses, vertex_probs, min_required_forces = self.toppling_model.predict(
            vertices, 
            normals, 
            push_directions, 
            use_sensitivity=self.use_sensitivity
        )

        from dexnet.visualization import DexNetVisualizer3D as vis3d
        from autolab_core import Point
        # j = 6 # gearbox
        j = 50 # vase
        a = j*self.toppling_model.n_trials
        b = (j+1)*self.toppling_model.n_trials
        for i in range(a,b):
            start = self.toppling_model.vertices[i]
            end = start - .01 * self.toppling_model.push_directions[i]
            vis3d.plot3d([start, end], color=[0,1,0], tube_radius=.0002)
        vis3d.mesh(state.mesh, state.T_obj_world)
        vis3d.show()
        return vertices, normals, poses, vertex_probs, min_required_forces
Example #3
0
def figure_0():
    action = policy.action(env.state)
    env.render_3d_scene()
    bottom_points = policy.toppling_model.bottom_points
    vis3d.plot3d(bottom_points[:2], color=[0, 0, 0], tube_radius=.001)

    mesh = env.state.mesh.copy().apply_transform(env.state.T_obj_world.matrix)
    mesh.fix_normals()
    direction = normalize([0, -.04, 0])
    origin = mesh.center_mass + np.array([0, .04, .09])
    intersect, _, face_ind = mesh.ray.intersects_location([origin],
                                                          [direction],
                                                          multiple_hits=False)
    normal = mesh.face_normals[face_ind[0]]
    start_point = intersect[0] + .06 * normal
    end_point = intersect[0]
    shaft_points = [start_point, end_point]
    h1 = np.array([[1, 0, 0], [0, 0.7071, -0.7071], [0, 0.7071,
                                                     0.7071]]).dot(-normal)
    h2 = np.array([[1, 0, 0], [0, 0.7071, 0.7071], [0, -0.7071,
                                                    0.7071]]).dot(-normal)
    head_points = [end_point - 0.02 * h2, end_point, end_point - 0.02 * h1]
    vis3d.plot3d(shaft_points, color=[1, 0, 0], tube_radius=.002)
    vis3d.plot3d(head_points, color=[1, 0, 0], tube_radius=.002)
    vis3d.points(Point(end_point), scale=.004, color=[0, 0, 0])
    hand_pose = RigidTransform(rotation=policy.get_hand_pose(
        start_point, end_point)[0].rotation,
                               translation=start_point,
                               from_frame='grasp',
                               to_frame='world')
    orig_pose = env.state.obj.T_obj_world.copy()
    env.state.obj.T_obj_world = policy.toppling_model.final_poses[1]
    action = policy.grasping_policy.action(env.state)
    env.state.obj.T_obj_world = orig_pose

    gripper = env.gripper(action)
    #vis3d.mesh(gripper.mesh, hand_pose * gripper.T_mesh_grasp, color=light_blue)
    # mesh = trimesh.load('~/Downloads/chris.stl')
    rot = np.array([[0, -1, 0], [1, 0, 0],
                    [0, 0,
                     1]]).dot(np.array([[0, 0, 1], [0, 1, 0], [-1, 0, 0]]))
    T = RigidTransform(rotation=rot,
                       translation=np.array([0, -.09, .1]),
                       to_frame='mesh',
                       from_frame='mesh')
    # vis3d.mesh(mesh, hand_pose * gripper.T_mesh_grasp * T, color=light_blue)
    vis3d.mesh(gripper.mesh,
               hand_pose * gripper.T_mesh_grasp * T,
               color=light_blue)
    vis3d.show(starting_camera_pose=CAMERA_POSE)

    env.state.obj.T_obj_world = policy.toppling_model.final_poses[1]
    action = policy.grasping_policy.action(env.state)
    print 'q:', action.q_value
    env.render_3d_scene()
    vis3d.gripper(env.gripper(action),
                  action.grasp(env.gripper(action)),
                  color=light_blue)
    vis3d.show(starting_camera_pose=CAMERA_POSE)
Example #4
0
    def antipodal_grasp_sampler(self):
        of = ObjFile(OBJ_FILENAME)
        sf = SdfFile(SDF_FILENAME)
        mesh = of.read()
        sdf = sf.read()
        obj = GraspableObject3D(sdf, mesh)

        gripper = RobotGripper.load(GRIPPER_NAME)

        ags = AntipodalGraspSampler(gripper, CONFIG)
        grasps = ags.generate_grasps(obj, target_num_grasps=10)

        quality_config = GraspQualityConfigFactory.create_config(
            CONFIG['metrics']['force_closure'])
        quality_fn = GraspQualityFunctionFactory.create_quality_function(
            obj, quality_config)

        q_to_c = lambda quality: CONFIG['quality_scale']

        i = 0
        vis.figure()
        vis.mesh(obj.mesh.trimesh, style='surface')
        for grasp in grasps:
            print(grasp.frame)
            success, c = grasp.close_fingers(obj)
            if success:
                c1, c2 = c
                fn_fc = quality_fn(grasp).quality
                true_fc = PointGraspMetrics3D.force_closure(
                    c1, c2, quality_config.friction_coef)
        #print(grasp)
            T_obj_world = RigidTransform(from_frame='obj', to_frame='world')
            color = plt.get_cmap('hsv')(q_to_c(CONFIG['metrics']))[:-1]
            T_obj_gripper = grasp.gripper_pose(gripper)
            #vis.grasp(grasp,grasp_axis_color=color,endpoint_color=color)
            i += 1
        #T_obj_world = RigidTransform(from_frame='obj',to_frame='world')
        vis.show(False)
Example #5
0
    def display_object(self, object_name, config=None):
        """Display an object
        
        Parameters
        ----------
        object_name : :obj:`str`
            Ob
            ject to display.
        config : :obj:`dict`
            Configuration dict for visualization.
            Parameters are in Other parameters. Values from self.default_config are used for keys not provided.
        
        Other Parameters
        ----------------
        animate
            Whether or not to animate the displayed object
            
        Raises
        ------
        ValueError
            invalid object name
        RuntimeError
            Database or dataset not opened.
        """
        self._check_opens()
        config = self._get_config(config)

        if object_name not in self.dataset.object_keys:
            raise ValueError(
                "{} is not a valid object name".format(object_name))

        logger.info('Displaying {}'.format(object_name))
        obj = self.dataset[object_name]

        vis.figure(bgcolor=(1, 1, 1), size=(1000, 1000))
        vis.mesh(obj.mesh, color=(0.5, 0.5, 0.5), style='surface')
        vis.show(animate=config['animate'])
Example #6
0
    def display_grasps(self,
                       object_name,
                       gripper_name,
                       metric_name,
                       config=None):
        """ Display grasps for an object 
        
        Parameters
        ----------
        object_name : :obj:`str`
            Object to display
        gripper_name : :obj:`str`
            Gripper for which to display grasps
        metric_name : :obj:`str`
            Metric to color/rank grasps with
        config : :obj:`dict`
            Configuration dict for visualization. 
            Parameters are in Other parameters. Values from self.default_config are used for keys not provided.
        
        Other Parameters
        ----------------
        gripper_dir
            Directory where the grippers models and parameters are.
        quality_scale
            Range to scale quality metric values to
        show_gripper
            Whether or not to show the gripper in the visualization
        min_metric
            lowest value of metric to show grasps for
        max_plot_gripper
            Number of grasps to plot
        animate
            Whether or not to animate the displayed object
        """
        self._check_opens()
        config = self._get_config(config)

        grippers = os.listdir(config['gripper_dir'])
        if gripper_name not in grippers:
            raise ValueError(
                "{} is not a valid gripper name".format(gripper_name))
        gripper = gr.RobotGripper.load(gripper_name,
                                       gripper_dir=config['gripper_dir'])

        objects = self.dataset.object_keys
        if object_name not in objects:
            raise ValueError(
                "{} is not a valid object name".format(object_name))

        metrics = self.dataset.available_metrics(object_name,
                                                 gripper=gripper.name)
        if metric_name not in metrics:
            raise ValueError(
                "{} is not computed for gripper {}, object {}".format(
                    metric_name, gripper.name, object_name))

        logger.info('Displaying grasps for gripper %s on object %s' %
                    (gripper.name, object_name))
        object = self.dataset[object_name]
        grasps, metrics = self.dataset.sorted_grasps(object_name,
                                                     metric_name,
                                                     gripper=gripper.name)

        if len(grasps) == 0:
            raise RuntimeError('No grasps for gripper %s on object %s' %
                               (gripper.name, object_name))
            return

        low = np.min(metrics)
        high = np.max(metrics)
        if low == high:
            q_to_c = lambda quality: config['quality_scale']
        else:
            q_to_c = lambda quality: config['quality_scale'] * (
                quality - low) / (high - low)

        if config['show_gripper']:
            i = 0
            stable_pose = self.dataset.stable_pose(object.key, 'pose_1')
            for grasp, metric in zip(grasps, metrics):
                if metric <= config['min_metric']:
                    continue

                print 'Grasp %d %s=%.5f' % (grasp.id, metric_name, metric)
                T_obj_world = RigidTransform(from_frame='obj',
                                             to_frame='world')
                color = plt.get_cmap('hsv')(q_to_c(metric))[:-1]
                T_obj_gripper = grasp.gripper_pose(gripper)
                grasp = grasp.perpendicular_table(stable_pose)
                vis.figure()
                vis.gripper_on_object(gripper,
                                      grasp,
                                      object,
                                      gripper_color=(0.25, 0.25, 0.25),
                                      stable_pose=stable_pose,
                                      plot_table=False)
                vis.show(animate=config['animate'])
                i += 1
                if i >= config['max_plot_gripper']:
                    break
        else:
            i = 0
            vis.figure()
            vis.mesh(object.mesh, style='surface')
            for grasp, metric in zip(grasps, metrics):
                if metric <= config['min_metric']:
                    continue

                print 'Grasp %d %s=%.5f' % (grasp.id, metric_name, metric)
                T_obj_world = RigidTransform(from_frame='obj',
                                             to_frame='world')
                color = plt.get_cmap('hsv')(q_to_c(metric))[:-1]
                T_obj_gripper = grasp.gripper_pose(gripper)
                vis.grasp(grasp, grasp_axis_color=color, endpoint_color=color)
                i += 1
                if i >= config['max_plot_gripper']:
                    break

            vis.show(animate=config['animate'])
Example #7
0
    env = GraspingEnv(config, config['vis'])
    env.reset()
    env.state.material_props._color = np.array([0.86274509803] * 3)
    obj_name = env.state.obj.key
    policy.set_environment(env.environment)
    if args.before:
        env.render_3d_scene()
        vis3d.show(starting_camera_pose=CAMERA_POSE)
    action = policy.action(env.state)

    vis3d.figure()
    env.render_3d_scene()
    num_vertices = len(action.metadata['vertices'])
    for i, vertex, q_increase in zip(np.arange(num_vertices),
                                     action.metadata['vertices'],
                                     action.metadata['quality_increases']):
        topples = action.metadata['final_pose_ind'][i] != 0
        color = np.array(
            [min(1, 2 * (1 - q_increase)),
             min(2 * q_increase, 1), 0]) if topples else np.array([0, 0, 0])
        # color = np.array([min(1, 2*(1-q_increase)), min(2*q_increase, 1), 0])
        scale = .001 if i != action.metadata['best_ind'] else .003
        vis3d.points(Point(vertex, 'world'), scale=scale, color=color)

    gripper = env.gripper(action)
    T_start_world = action.T_begin_world * gripper.T_mesh_grasp
    T_end_world = action.T_end_world * gripper.T_mesh_grasp
    vis3d.mesh(gripper.mesh, T_start_world, color=(0, 1, 0))
    vis3d.mesh(gripper.mesh, T_end_world, color=(1, 0, 0))
    display_or_save('{}_quality_increases.gif'.format(obj_name))
Example #8
0
        env.render_3d_scene()
        vis3d.show(starting_camera_pose=CAMERA_POSE)

        obj_config = config['state_space']['object']
        all_poses, _ = env.state.obj.mesh.compute_stable_poses(
            sigma=obj_config['stp_com_sigma'],
            n_samples=obj_config['stp_num_samples'],
            threshold=.001)
        # for pose in all_poses:
        #     env.state.obj.T_obj_world = to_rigid(pose)
        #     env.render_3d_scene()
        #     vis3d.show(starting_camera_pose=CAMERA_POSE)

        mesh = trimesh.load('~/Downloads/chris.stl')
        vis3d.mesh(mesh)
        vis3d.show()

    #figure_1()
    #figure_2()
    #figure_3()
    #figure_0()
    # figure_3()
    # figure_0()
    #failure_modes()
    action = policy.action(env.state)
    #noise_vis()

    if False:
        j = np.argmax(action.metadata['vertex_probs'][:, 2])
        R = policy.toppling_model.tipping_point_rotations()[1]
Example #9
0
def render_3d_scene(env):
    vis3d.mesh(env.state.mesh,
               env.state.T_obj_world.matrix,
               color=env.state.color)
Example #10
0
    def showgraspfrompicklewithcandidate(self, object_name,metric_name,onlynagative,onlypositive,openravechecker):    
        object = self.dataset[object_name] 
        labelgrasps,labelmetrics,candidate_labelgrasps,candidate_labelmetrics=self.readdatafrompicklewithcandidate(object_name)


        config=self._get_config(None)
        low =0.0  #np.min(metrics)
        high = np.max([labelmetrics[i][metric_name] for i in range(len(labelmetrics))])
       # print 'high',high,config['quality_scale']
        if low == high:
            q_to_c = lambda quality: config['quality_scale']
        else:
            
            q_to_c = lambda quality: config['quality_scale'] * (quality - low) / (high - low)

        recalculate_grasp=[]    
        vis.figure()
        vis.mesh(object.mesh.trimesh ,style='surface')
        for i in range(len(labelgrasps)): 
         
                        if labelmetrics[i][metric_name]==0 : 
                           if not onlypositive: 
                                    vis.figure()
                                    vis.mesh(object.mesh.trimesh ,style='surface')
                                    color = plt.get_cmap('hsv')(q_to_c(labelmetrics[i][metric_name]))[:-1] 
                                    
                                    vis.graspwithapproachvectorusingcenter_point(labelgrasps[i] ,approaching_color=(1,0,0), grasp_axis_color=(1,0,0) )
                                    for kk in range(len(candidate_labelgrasps[i] )):  
                                        print 'candidate_labelgrasps[i][kk]',candidate_labelgrasps[i][kk]
                                        color = plt.get_cmap('hsv')(q_to_c(candidate_labelmetrics[i][candidate_labelgrasps[i][kk].id][metric_name]))[:-1] 
                                        vis.graspwithapproachvectorusingcenter_point(candidate_labelgrasps[i][kk] ,approaching_color=color, grasp_axis_color=color )
                         
                                    recalculate_grasp.append(labelgrasps[i])
                                    print 'maxtrics',i,labelmetrics[i][metric_name]   # ,high
                                
                                    vis.pose(RigidTransform(), alpha=0.1) 
                    
                                    vis.show(animate=False)             
                        elif  labelmetrics[i][metric_name]==-1   :
                            if not onlypositive:
                                 vis.shownormals(labelgrasps[i][1],labelgrasps[i][0],color=(0,0,1) ) 
                        else:
                                if not onlynagative: 
                                    vis.figure()
                                    vis.mesh(object.mesh.trimesh ,style='surface')
                                    color = plt.get_cmap('hsv')(q_to_c(labelmetrics[i][metric_name]))[:-1]                               
                                    if len(candidate_labelgrasps[i])>0:
                                        A=np.array(labelgrasps[i].rotated_full_axis[:,0] )
                                        B=np.array(candidate_labelgrasps[i][0].rotated_full_axis[:,0] ) 
                                        num = np.dot(A.T, B)
                                        print num
                                        if num<0.99:
                                            labelgrasps[i].approach_angle_= labelgrasps[i]._angle_aligned_with_table( -candidate_labelgrasps[i][0].rotated_full_axis[:,0])
                                                                        
                                    vis.graspwithapproachvectorusingcenter_point(labelgrasps[i] ,approaching_color=color, grasp_axis_color=(0,0,1) )
                                    print 'length',len(candidate_labelgrasps[i] )
                                    for kk in range(len(candidate_labelgrasps[i] )):   
                                        color = plt.get_cmap('hsv')(q_to_c(candidate_labelmetrics[i][candidate_labelgrasps[i][kk].id][metric_name]))[:-1] 
                                        vis.graspwithapproachvectorusingcenter_point(candidate_labelgrasps[i][kk] ,approaching_color=(0,1,1), grasp_axis_color=color )
 
                                    recalculate_grasp.append(labelgrasps[i])
                                    print 'maxtrics',i,labelmetrics[i][metric_name]   # ,high
                                
                                    vis.pose(RigidTransform(), alpha=0.1) 
                                    vis.show(animate=False) 
Example #11
0
    def antipodal_grasp_sampler(self):
    	#of = ObjFile(OBJ_FILENAME)
    	#sf = SdfFile(SDF_FILENAME)
    	#mesh = of.read()
    	#sdf = sf.read()
        #obj = GraspableObject3D(sdf, mesh)
        mass = 1.0
        CONFIG['obj_rescaling_type'] = RescalingType.RELATIVE
        mesh_processor = MeshProcessor(OBJ_FILENAME, CONFIG['cache_dir'])
        mesh_processor.generate_graspable(CONFIG)
        mesh = mesh_processor.mesh
        sdf = mesh_processor.sdf
        obj = GraspableObject3D(sdf, mesh)

        gripper = RobotGripper.load(GRIPPER_NAME)

        ags = AntipodalGraspSampler(gripper, CONFIG)
        grasps = ags.sample_grasps(obj, num_grasps=100)
        print(len(grasps))
  
        quality_config = GraspQualityConfigFactory.create_config(CONFIG['metrics']['robust_ferrari_canny'])
        #quality_config = GraspQualityConfigFactory.create_config(CONFIG['metrics']['force_closure'])
        quality_fn = GraspQualityFunctionFactory.create_quality_function(obj, quality_config)
        

        vis.figure()
        #vis.points(PointCloud(nparraypoints), scale=0.001, color=np.array([0.5,0.5,0.5]))
        #vis.plot3d(nparraypoints)

        metrics = []
        vis.mesh(obj.mesh.trimesh, style='surface')
        for grasp in grasps:
            success, c = grasp.close_fingers(obj)
            if success:
                c1, c2 = c
                #fn_fc = quality_fn(grasp).quality
                true_fc = PointGraspMetrics3D.grasp_quality(grasp, obj, quality_config)
                true_fc = true_fc
                metrics.append(true_fc)

    	    #color = quality_fn(grasp).quality
            #if (true_fc > 1.0):
            #    true_fc = 1.0
            #color = (1.0-true_fc, true_fc, 0.0)

        low = np.min(metrics)
        high = np.max(metrics)
        print(low)
        print(high)
        if low == high:
            q_to_c = lambda quality: CONFIG['quality_scale']
        else:
            q_to_c = lambda quality: CONFIG['quality_scale'] * (quality - low) / (high - low)
        print(len(metrics))
        for grasp, metric in zip(grasps, metrics):
            color = plt.get_cmap('hsv')(q_to_c(metric))[:-1]
            print(color)
            vis.grasp(grasp, grasp_axis_color=color,endpoint_color=color)


        vis.show(False)