CADMatch

Introduction

The CADMatch module is an optional module of the rc_visard NG and requires a separate CADMatch license to be purchased.

This module provides an out-of-the-box perception solution for 3D object detection and grasping. CADMatch targets the detection of 3D objects based on a CAD template for picking with a general gripper. The objects can be located in a bin or placed arbitrarily in the field of view of the camera.

For the CADMatch module to work, special object templates are required for each type of object to be detected. Please get in touch with the Roboception support (Contact) to order a template for your CAD file.

The CADMatch module offers:

  • A dedicated page on the rc_visard NG Web GUI for easy setup, configuration, testing, and application tuning.
  • A REST-API interface and a KUKA Ethernet KRL Interface.
  • The definition of regions of interest to select relevant volumes in the scene (see RoiDB).
  • A load carrier detection functionality for bin-picking applications (see LoadCarrier), to provide grasps for objects inside a bin only.
  • The definition of compartments inside a load carrier to provide grasps for specific volumes of the bin only.
  • The option to use user-defined object pose priors.
  • Storing of up to 50 templates.
  • The definition of up to 100 grasp points for each template via an interactive visualization in the Web GUI.
  • Collision checking between the gripper and the load carrier, other detected objects and/or the point cloud.
  • Collision checking between the object in the gripper and the load carrier walls during retraction.
  • Support for static and robot-mounted cameras and optional integration with the Hand-eye calibration module, to provide grasps in the user-configured external reference frame.
  • Selection of a sorting strategy to sort the detected objects and returned grasps.
  • 3D visualization of the detection results with grasp points and gripper animations in the Web GUI.

Setting of grasp points

The CADMatch module detects 3D objects in a scene based on a CAD template and returns the poses of the object origins. To use CADMatch directly in a robot application, up to 100 grasp points can be defined for each template. A grasp point represents the desired position and orientation of the robot’s TCP (Tool Center Point) to grasp an object.

Please consult Setting of grasp points for further details.

Setting grasp points in the Web GUI

The rc_visard NG Web GUI provides an intuitive and interactive way of defining grasp points for object templates. In a first step, the object template has to be uploaded to the rc_visard NG. This can be done in the Web GUI in any pipeline under Modules ‣ CADMatch by clicking on + Add a new Template in the Templates, Grasps and Pose Priors section, or in Database ‣ Templates in the CADMatch Templates, Grasps and Pose Priors section. Once the template upload is complete, a dialog with a 3D visualization of the object template is shown for adding or editing grasp points. The same dialog appears when editing an existing template.

More details are given in Setting grasp points in the Web GUI.

Setting grasp points via the REST-API

Grasp points can be set via the REST-API interface using the set_grasp or set_all_grasps services (see Internal services).

More details are given in Setting grasp points via the REST-API.

Setting of pose priors

The CADMatch module offers the possibility to define prior poses of the objects to be detected. If a pose prior is given, the object detection will use this pose prior and only refine the given pose. This speeds up the detection significantly. A pose prior represents the approximate position and orientation of the object to be detected. The pose can be defined in the camera or the external coordinate frame, if a hand-eye calibration is available.

Each pose prior consists of an id which must be unique within all pose priors for an object template, the template_id representing the template the pose prior applies to, the pose and the pose_frame of the prior. Pose priors can be set via the REST-API interface, or by using the interactive visualization in the Web GUI. The Web GUI allows to interactively position the object in the current point cloud. This can be done in the “Pose Priors” tab during editing a template.

Pose priors should be used in applications where the approximate object poses are known beforehand. The rc_visard NG can store up to 50 pose priors per template.

Setting the preferred orientation of the TCP

The CADMatch module determines the reachability of grasp points based on the preferred orientation of the gripper or TCP. The preferred orientation can be set via the set_preferred_orientation service call or on the CADMatch page in the Web GUI. The resulting direction of the TCP’s z axis is used to reject grasps which cannot be reached by the gripper. Furthermore, the preferred orientation can be used to sort the reachable grasps by setting the corresponding sorting strategy.

The preferred orientation can be set in the camera coordinate frame or in the external coordinate frame, in case a hand-eye calibration is available. If the preferred orientation is specified in the external coordinate frame and the sensor is robot mounted, the current robot pose has to be given to each object detection call, so that the preferred orientation can be used for filtering and, optionally, sorting the grasps on the detected objects. If no preferred orientation is set, the orientation of the left camera is used as the preferred orientation of the TCP.

Setting the sorting strategies

The objects and grasps returned by the detect_object service call are sorted according to a sorting strategy which can be chosen by the user. The following sorting strategies are available and can be set in the Web GUI or using the set_sorting_strategies service call:

  • gravity: highest matches and grasp points along the gravity direction are returned first,
  • match_score: matches with the highest match score and grasp points on objects with the highest match score are returned first,
  • preferred_orientation: matches and grasp points with minimal rotation difference between their orientation and the preferred orientation of the TCP are returned first,
  • direction: matches and grasp points with the shortest distance along a defined direction vector in a given pose_frame are returned first.

If no sorting strategy is set or default sorting is chosen in the Web GUI, sorting is done based on a combination of match_score and the minimal distance from the camera along the z axis of the preferred orientation of the TCP.

Detection of objects

The CADMatch module requires an object template for object detection. This template contains information about the 3D shape of the object and prominent edges that can be visible in the camera images. CADMatch also supports partial object templates, which contain only a specific part of the object that can be detected well, e.g., in case of occlusions. Furthermore, templates can require a pose prior for the detection which is then only refined using the image data.

The object detection is a two-stage process consisting of a prior estimation step and a pose refinement step. First, a pose prior is computed based on the appearance of the object in the camera images. Second, the pose is refined by using the 3D point cloud and edges in the camera image. For this to work, the objects to detect must be visible in both left and right camera images. If pose priors are given, only the pose refinement step is performed based, which decreases runtime significantly.

For triggering the object detection, in general, the following information must be provided to the CADMatch module:

  • The template ID of the object to be detected in the scene.
  • The coordinate frame in which the poses of the detected objects and the grasp points shall be returned (ref. Hand-eye calibration).

Optionally, further information can be given to the CADMatch module:

  • The IDs of the pose priors which approximately match the poses of the objects to be detected. In case a template is used that requires a pose prior, one or more pose prior IDs have to be provided.
  • The ID of the load carrier which contains the items to be detected.
  • A compartment inside the load carrier where to detect objects (see Load carrier compartments).
  • The ID of the 3D region of interest where to search for the load carriers if a load carrier is set. Otherwise, the ID of the 3D region of interest where to search for the objects.
  • The current robot pose in case the camera is mounted on the robot and the chosen coordinate frame for the poses is external, or the preferred orientation is given in the external frame, or the chosen region of interest is defined in the external frame.
  • Collision detection information: The ID of the gripper to enable collision checking and optionally a pre-grasp offset to define a pre-grasp position. Details on collision checking are given below in CollisionCheck.
  • Data acquisition mode: The user can choose if a new image dataset is acquired for the detection (default), or if the detection should be performed on the previously used image dataset. This saves data acquisition time, e.g. in case several detections with different templates have to be run on the same image.

On the Web GUI the detection can be tested in the Try Out section of the CADMatch module’s page.

The detected objects are returned in a list of matches, sorted according to the selected sorting strategy (see Setting the sorting strategies). Each detected object includes a uuid (Universally Unique Identifier) and the timestamp of the oldest image that was used to detect it. The pose of a detected object corresponds to the pose of the origin of the object template used for detection. Furthermore, the matching score is given to indicate the quality of the detection.

If the chosen template also has grasp points attached (see Setting of grasp points), a list of grasps for all objects is returned in addition to the list of detected objects. The grasps are sorted according to the selected sorting strategy (see Setting the sorting strategies). The grasp poses are given in the desired coordinate frame. There are references between the detected objects and the grasps via their uuids.

For objects with a discrete symmetry (e.g. prismatic objects), all collision-free symmetries of each grasp point which are reachable according to the given preferred TCP orientation are returned, ordered by the given sorting strategy.

For objects with a continuous symmetry (e.g. cylindrical objects), all grasps symmetric to each grasp point on an object are checked for reachability and collisions, and only the best one according to the given sorting strategy is returned.

Note

The first detection call with a new object template takes longer than the following detection calls, because the object template has to be loaded into the CADMatch module first. To avoid this, the warmup_template service can be used to load a template so that it is ready when the first detection is triggered.

Interaction with other modules

Internally, the CADMatch module depends on, and interacts with other on-board modules as listed below.

Note

All changes and configuration updates to these modules will affect the performance of the CADMatch modules.

Stereo camera and Stereo matching

The CADMatch module makes internally use of the following data:

  • Rectified images from the Camera module (rc_camera);
  • Disparity, error, and confidence images from the Stereo matching module (rc_stereomatching).

The quality parameter of the stereo matching module must be set to Medium or higher (see Parameters). We recommend Full or High quality for using CADMatch.

All processed images are guaranteed to be captured after the module trigger time.

IO and Projector Control

In case the rc_visard NG is used in conjunction with an external random dot projector and the IO and Projector Control module (rc_iocontrol), it is recommended to connect the projector to GPIO Out 1 and set the stereo-camera module’s acquisition mode to SingleFrameOut1 (see Stereo matching parameters), so that on each image acquisition trigger an image with and without projector pattern is acquired.

Alternatively, the output mode for the GPIO output in use should be set to ExposureAlternateActive (see Description of run-time parameters).

In either case, the Auto Exposure Mode exp_auto_mode should be set to AdaptiveOut1 to optimize the exposure of both images (see Stereo camera parameters).

Hand-eye calibration

In case the camera has been calibrated to a robot, the CADMatch module can automatically provide poses in the robot coordinate frame. For the CADMatch node’s Services, the frame of the output poses can be controlled with the pose_frame argument.

Two different pose_frame values can be chosen:

  1. Camera frame (camera). All poses provided by the modules are in the camera frame, and no prior knowledge about the pose of the camera in the environment is required. This means that the configured regions of interest and load carriers move with the camera. It is the user’s responsibility to update the configured poses if the camera frame moves (e.g. with a robot-mounted camera).
  2. External frame (external). All poses provided by the modules are in the external frame, configured by the user during the hand-eye calibration process. The module relies on the on-board Hand-eye calibration module to retrieve the sensor mounting (static or robot mounted) and the hand-eye transformation. If the mounting is static, no further information is needed. If the sensor is robot-mounted, the robot_pose is required to transform poses to and from the external frame.

Note

If no hand-eye calibration is available, all pose_frame values should be set to camera.

All pose_frame values that are not camera or external are rejected.

If the sensor is robot-mounted, the current robot_pose has to be provided depending on the value of pose_frame, the definition of the preferred TCP orientation and the sorting direction:

  • If pose_frame is set to external, providing the robot pose is obligatory.
  • If the preferred TCP orientation is defined in external, providing the robot pose is obligatory.
  • If the sorting direction is defined in external, providing the robot pose is obligatory.
  • In all other cases, providing the robot pose is optional.

LoadCarrier

The CADMatch module uses the load carrier detection functionality provided by the LoadCarrier module (rc_load_carrier), with the run-time parameters specified for this module. However, only one load carrier will be returned and used in case multiple matching load carriers could be found in the scene. In case multiple load carriers of the same type are visible, a region of interest should be set to ensure that always the same load carrier is used for the CADMatch module.

CollisionCheck

Collision checking can be easily enabled for grasp computation of the CADMatch module by passing a collision_detection argument to the detect_object service call. It contains the ID of the used gripper and optionally a pre-grasp offset. The gripper has to be defined in the GripperDB module (see Setting a gripper) and details about collision checking are given in Collision checking within other modules.

Alternatively, grasp points can be assigned individual gripper IDs, and collision checking can be enabled for all grasp points with gripper IDs by enabling the run-time parameter check_collisions.

If the selected CADMatch template contains a collision geometry and the run-time parameter check_collisions_with_matches is true, also collisions between the gripper and all other detected objects (not limited to max_matches) will be checked. The object on which the grasp point to be checked is located, is excluded from the collision check.

If the run-time parameter check_collisions_with_point_cloud is true, also collisions between the gripper and a watertight version of the point cloud are checked. If this feature is used with suctions grippers, it should be ensured that the TCP is defined to be outside the gripper geometry, or that the grasp points are defined above the object surface. Otherwise every grasp will result in a collision between the gripper and the point cloud.

If the run-time parameter check_collisions_during_retraction is true and a load carrier and a pre-grasp offset are given, each grasp point will be checked for collisions between the object in the gripper and the load carrier walls during retraction. This collision check is performed along the full linear trajectory from the grasp point back to the pre-grasp position.

If collision checking is enabled, only grasps which are collision free or could not be checked for collisions (e.g. because no gripper was given) will be returned. The result image on top of the CADMatch page of the Web GUI also shows collision-free grasps in green, unchecked grasps in yellow and colliding grasp points in red. The detected objects which are considered in the collision check are also visualized with their edges in red.

The CollisionCheck module’s run-time parameters affect the collision detection as described in CollisionCheck Parameters.

Parameters

The CADMatch module is called rc_cadmatch in the REST-API and is represented in the Web GUI under Modules ‣ CADMatch. The user can explore and configure the rc_cadmatch module’s run-time parameters, e.g. for development and testing, using the Web GUI or the REST-API interface.

Parameter overview

This module offers the following run-time parameters:

The rc_cadmatch module’s run-time parameters
Name Type Min Max Default Description
check_collisions bool false true false Whether to check for collisions when a gripper is defined for a grasp
check_collisions_-during_retraction bool false true false Whether to check for collisions between the object in the gripper and the load carrier during retraction
check_collisions_with_matches bool false true true Whether to check for collisions between gripper and detected matches
check_collisions_-with_point_cloud bool false true false Whether to check for collisions between gripper and point cloud
edge_max_distance float64 0.5 5.0 2.0 Maximum allowed distance in pixels between the template edges and the detected edges in the image
edge_sensitivity float64 0.05 1.0 0.5 Sensitivity of the edge detector
grasp_filter_-orientation_threshold float64 0.0 180.0 45.0 Maximum allowed orientation change between grasp and preferred orientation in degrees
max_matches int32 1 30 10 Maximum number of matches
min_score float64 0.05 1.0 0.3 Minimum score for matches
only_highest_priority_grasps bool false true false Whether to return only the highest priority level grasps

Description of run-time parameters

Each run-time parameter is represented by a row on the Web GUI’s CADMatch page. The name in the Web GUI is given in brackets behind the parameter name and the parameters are listed in the order they appear in the Web GUI:

max_matches (Maximum Matches)

is the maximum number of objects to detect.

Via the REST-API, this parameter can be set as follows.

PUT http://<host>/api/v2/pipelines/<0,1,2,3>/nodes/rc_cadmatch/parameters?max_matches=<value>
PUT http://<host>/api/v1/nodes/rc_cadmatch/parameters?max_matches=<value>

min_score (Minimum Score)

is the minimum detection score after refinement. The higher this value, the better 2D edges and 3D point cloud must match the given template.

Via the REST-API, this parameter can be set as follows.

PUT http://<host>/api/v2/pipelines/<0,1,2,3>/nodes/rc_cadmatch/parameters?min_score=<value>
PUT http://<host>/api/v1/nodes/rc_cadmatch/parameters?min_score=<value>

edge_sensitivity (Edge Sensitivity)

is the sensitivity of the edge detector. The higher the value of this parameter, the more edges will be used for pose refinement.

Via the REST-API, this parameter can be set as follows.

PUT http://<host>/api/v2/pipelines/<0,1,2,3>/nodes/rc_cadmatch/parameters?edge_sensitivity=<value>
PUT http://<host>/api/v1/nodes/rc_cadmatch/parameters?edge_sensitivity=<value>

edge_max_distance (Maximum Edge Distance)

is the maximum allowed distance in pixels between the template edges and the detected edges in the image during the refinement step.

Via the REST-API, this parameter can be set as follows.

PUT http://<host>/api/v2/pipelines/<0,1,2,3>/nodes/rc_cadmatch/parameters?edge_max_distance=<value>
PUT http://<host>/api/v1/nodes/rc_cadmatch/parameters?edge_max_distance=<value>

grasp_filter_orientation_threshold (Grasp Orientation Threshold)

is the maximum deviation of the TCP’s z axis at the grasp point from the z axis of the TCP’s preferred orientation in degrees. Only grasp points which are within this threshold are returned. When set to zero, any deviations are valid.

Via the REST-API, this parameter can be set as follows.

PUT http://<host>/api/v2/pipelines/<0,1,2,3>/nodes/rc_cadmatch/parameters?grasp_filter_orientation_threshold=<value>
PUT http://<host>/api/v1/nodes/rc_cadmatch/parameters?grasp_filter_orientation_threshold=<value>

only_highest_priority_grasps (Only Highest Priority Grasps)

If set to true, only grasps with the highest priority will be returned. If collision checking is enabled, only the collision-free grasps among the group of grasps with the highest priority are returned. This can save computation time and reduce the number of grasps to be parsed on the application side.

Without collision checking, only grasps of highest priority are returned.

PUT http://<host>/api/v2/pipelines/<0,1,2,3>/nodes/rc_cadmatch/parameters?only_highest_priority_grasps=<value>
PUT http://<host>/api/v1/nodes/rc_cadmatch/parameters?only_highest_priority_grasps=<value>

check_collisions (Check Collisions)

If this parameter is enabled, collision checking will be performed for all grasps which have a gripper ID assigned, even when no default gripper is given in the detect_object service call. If a load carrier is used, the collision check will always be performed between the gripper and the load carrier. Collision checking with the point cloud and other matches is only performed when the corresponding runtime parameters are enabled.

Via the REST-API, this parameter can be set as follows.

PUT http://<host>/api/v2/pipelines/<0,1,2,3>/nodes/rc_cadmatch/parameters?check_collisions=<value>
PUT http://<host>/api/v1/nodes/rc_cadmatch/parameters?check_collisions=<value>

check_collisions_with_matches (Check Collisions with Matches)

This parameter is only used when collision checking is enabled by passing a gripper to the detect_object service call or by enabling the check_collisions runtime parameter. If check_collisions_with_matches is set to true, all grasp points will be checked for collisions between the gripper and all other detected objects (not limited to max_matches), and only grasp points at which the gripper would not collide with any other detected object will be returned.

Via the REST-API, this parameter can be set as follows.

PUT http://<host>/api/v2/pipelines/<0,1,2,3>/nodes/rc_cadmatch/parameters?check_collisions_with_matches=<value>
PUT http://<host>/api/v1/nodes/rc_cadmatch/parameters?check_collisions_with_matches=<value>

check_collisions_with_point_cloud (Check Collisions with Point Cloud)

This parameter is only used when collision checking is enabled by passing a gripper to the detect_object service call or by enabling the check_collisions runtime parameter. If check_collisions_with_point_cloud is set to true, all grasp points will be checked for collisions between the gripper a watertight version of the point cloud, and only grasp points at which the gripper would not collide with this point cloud will be returned.

Via the REST-API, this parameter can be set as follows.

PUT http://<host>/api/v2/pipelines/<0,1,2,3>/nodes/rc_cadmatch/parameters?check_collisions_with_point_cloud=<value>
PUT http://<host>/api/v1/nodes/rc_cadmatch/parameters?check_collisions_with_point_cloud=<value>

check_collisions_during_retraction (Check Collisions during Retraction)

This parameter is only used when collision checking is enabled by passing a gripper to the detect_object service call or by enabling the check_collisions runtime parameter. When check_collisions_during_retraction is enabled and a load carrier and a pre-grasp offset are given, each grasp point will be checked for collisions between the object in the gripper and the load carrier walls during retraction. This collision checking is performed along the full linear trajectory from the grasp point back to the pre-grasp position. Only collision-free grasp points will be returned.

Via the REST-API, this parameter can be set as follows.

PUT http://<host>/api/v2/pipelines/<0,1,2,3>/nodes/rc_cadmatch/parameters?check_collisions_during_retraction=<value>
PUT http://<host>/api/v1/nodes/rc_cadmatch/parameters?check_collisions_during_retraction=<value>

Status values

The rc_cadmatch module reports the following status values:

The rc_cadmatch module’s status values
Name Description
data_acquisition_time Time in seconds required by the last active service to acquire images
last_timestamp_processed The timestamp of the last processed dataset
last_request_timestamp The timestamp of the last detection request
load_carrier_detection_time Processing time of the last load carrier detection in seconds
object_detection_time Processing time of the last last object detection in seconds
processing_time Processing time of the last detection (including load carrier detection) in seconds
state The current state of the rc_cadmatch node

The reported state can take one of the following values.

Possible states of the CADMatch module
State name Description
IDLE The module is idle.
RUNNING The module is running and ready for load carrier detection and object detection.
FATAL A fatal error has occurred.

Services

The user can explore and call the rc_cadmatch module’s services, e.g. for development and testing, using the REST-API interface or the rc_visard NG Web GUI.

The CADMatch modules offer the following services.

detect_object

Triggers the detection of objects as described in Detection of objects based on an object template.

Details

This service can be called as follows.

PUT http://<host>/api/v2/pipelines/0/nodes/rc_cadmatch/services/detect_object
PUT http://<host>/api/v1/nodes/rc_cadmatch/services/detect_object

Required arguments:

pose_frame: see Hand-eye calibration.

template_id: the ID of the template to be detected.

Potentially required arguments:

robot_pose: see Hand-eye calibration.

pose_prior_ids: IDs of the pose priors for the items to be detected. In case the chosen template requires a pose prior for the detection, this argument must be provided.

Optional arguments:

load_carrier_id: ID of the load carrier which contains the items to be detected.

load_carrier_compartment: compartment inside the load carrier where to detect items (see Load carrier compartments).

region_of_interest_id: if load_carrier_id is set, ID of the 3D region of interest where to search for the load carriers. Otherwise, ID of the 3D region of interest where to search for the objects.

collision_detection: see Collision checking within other modules.

data_acquisition_mode: if set to CAPTURE_NEW (default), a new image dataset will be used for the detection. If set to USE_LAST the previous dataset will be used for the detection.

The definition for the request arguments with corresponding datatypes is:

{
  "args": {
    "collision_detection": {
      "gripper_id": "string",
      "pre_grasp_offset": {
        "x": "float64",
        "y": "float64",
        "z": "float64"
      }
    },
    "data_acquisition_mode": "string",
    "load_carrier_compartment": {
      "box": {
        "x": "float64",
        "y": "float64",
        "z": "float64"
      },
      "pose": {
        "orientation": {
          "w": "float64",
          "x": "float64",
          "y": "float64",
          "z": "float64"
        },
        "position": {
          "x": "float64",
          "y": "float64",
          "z": "float64"
        }
      }
    },
    "load_carrier_id": "string",
    "pose_frame": "string",
    "pose_prior_ids": [
      "string"
    ],
    "region_of_interest_id": "string",
    "robot_pose": {
      "orientation": {
        "w": "float64",
        "x": "float64",
        "y": "float64",
        "z": "float64"
      },
      "position": {
        "x": "float64",
        "y": "float64",
        "z": "float64"
      }
    },
    "template_id": "string"
  }
}

grasps: list of grasps on the detected objects, ordered according to the chosen sorting strategy. The match_uuid gives the reference to the detected object in matches this grasp belongs to. The list of returned grasps will be trimmed to the 100 best grasps if more reachable grasps are found. Each grasp contains a flag collision_checked and a gripper_id (see Collision checking within other modules).

load_carriers: list of detected load carriers.

matches: list of detected objects matching the template. The matches are ordered according to the chosen sorting strategy. The score indicates how well the object matches the template. The grasp_uuids refer to the grasps in grasps which are reachable on this object.

timestamp: timestamp of the image set the detection ran on.

return_code: holds possible warnings or error codes and messages.

The definition for the response with corresponding datatypes is:

{
  "name": "detect_object",
  "response": {
    "grasps": [
      {
        "collision_checked": "bool",
        "gripper_id": "string",
        "id": "string",
        "match_uuid": "string",
        "pose": {
          "orientation": {
            "w": "float64",
            "x": "float64",
            "y": "float64",
            "z": "float64"
          },
          "position": {
            "x": "float64",
            "y": "float64",
            "z": "float64"
          }
        },
        "pose_frame": "string",
        "priority": "int8",
        "timestamp": {
          "nsec": "int32",
          "sec": "int32"
        },
        "uuid": "string"
      }
    ],
    "load_carriers": [
      {
        "height_open_side": "float64",
        "id": "string",
        "inner_dimensions": {
          "x": "float64",
          "y": "float64",
          "z": "float64"
        },
        "outer_dimensions": {
          "x": "float64",
          "y": "float64",
          "z": "float64"
        },
        "overfilled": "bool",
        "pose": {
          "orientation": {
            "w": "float64",
            "x": "float64",
            "y": "float64",
            "z": "float64"
          },
          "position": {
            "x": "float64",
            "y": "float64",
            "z": "float64"
          }
        },
        "pose_frame": "string",
        "rim_ledge": {
          "x": "float64",
          "y": "float64"
        },
        "rim_step_height": "float64",
        "rim_thickness": {
          "x": "float64",
          "y": "float64"
        },
        "type": "string"
      }
    ],
    "matches": [
      {
        "grasp_uuids": [
          "string"
        ],
        "pose": {
          "orientation": {
            "w": "float64",
            "x": "float64",
            "y": "float64",
            "z": "float64"
          },
          "position": {
            "x": "float64",
            "y": "float64",
            "z": "float64"
          }
        },
        "pose_frame": "string",
        "score": "float32",
        "template_id": "string",
        "timestamp": {
          "nsec": "int32",
          "sec": "int32"
        },
        "uuid": "string"
      }
    ],
    "return_code": {
      "message": "string",
      "value": "int16"
    },
    "timestamp": {
      "nsec": "int32",
      "sec": "int32"
    }
  }
}

set_preferred_orientation

Persistently stores the preferred orientation of the gripper to compute the reachability of the grasps, which is used for filtering and, optionally, sorting the grasps returned by the detect_object service (see Setting the preferred orientation of the TCP).

Details

This service can be called as follows.

PUT http://<host>/api/v2/pipelines/0/nodes/rc_cadmatch/services/set_preferred_orientation
PUT http://<host>/api/v1/nodes/rc_cadmatch/services/set_preferred_orientation

The definition for the request arguments with corresponding datatypes is:

{
  "args": {
    "orientation": {
      "w": "float64",
      "x": "float64",
      "y": "float64",
      "z": "float64"
    },
    "pose_frame": "string"
  }
}

The definition for the response with corresponding datatypes is:

{
  "name": "set_preferred_orientation",
  "response": {
    "return_code": {
      "message": "string",
      "value": "int16"
    }
  }
}

get_preferred_orientation

Returns the preferred orientation of the gripper to compute the reachability of the grasps, which is used for filtering and, optionally, sorting the grasps returned by the detect_object service (see Setting the preferred orientation of the TCP).

Details

This service can be called as follows.

PUT http://<host>/api/v2/pipelines/0/nodes/rc_cadmatch/services/get_preferred_orientation
PUT http://<host>/api/v1/nodes/rc_cadmatch/services/get_preferred_orientation
This service has no arguments.

The definition for the response with corresponding datatypes is:

{
  "name": "get_preferred_orientation",
  "response": {
    "orientation": {
      "w": "float64",
      "x": "float64",
      "y": "float64",
      "z": "float64"
    },
    "pose_frame": "string",
    "return_code": {
      "message": "string",
      "value": "int16"
    }
  }
}

set_sorting_strategies

Persistently stores the sorting strategy for sorting the grasps and matches returned by the detect_object service (see Detection of objects).

Details

This service can be called as follows.

PUT http://<host>/api/v2/pipelines/0/nodes/rc_cadmatch/services/set_sorting_strategies
PUT http://<host>/api/v1/nodes/rc_cadmatch/services/set_sorting_strategies

Only one strategy may have a weight greater than 0. If all weight values are set to 0, the module will use the default sorting strategy.

If the weight for direction is set, the vector must contain the direction vector and pose_frame must be either camera or external.

The definition for the request arguments with corresponding datatypes is:

{
  "args": {
    "direction": {
      "pose_frame": "string",
      "vector": {
        "x": "float64",
        "y": "float64",
        "z": "float64"
      },
      "weight": "float64"
    },
    "gravity": {
      "weight": "float64"
    },
    "match_score": {
      "weight": "float64"
    },
    "preferred_orientation": {
      "weight": "float64"
    }
  }
}

The definition for the response with corresponding datatypes is:

{
  "name": "set_sorting_strategies",
  "response": {
    "return_code": {
      "message": "string",
      "value": "int16"
    }
  }
}

get_sorting_strategies

Returns the sorting strategy for sorting the grasps and matches returned by the detect_object service (see Detection of objects).

Details

This service can be called as follows.

PUT http://<host>/api/v2/pipelines/0/nodes/rc_cadmatch/services/get_sorting_strategies
PUT http://<host>/api/v1/nodes/rc_cadmatch/services/get_sorting_strategies
This service has no arguments.

All weight values are 0 when the module uses the default sorting strategy.

The definition for the response with corresponding datatypes is:

{
  "name": "get_sorting_strategies",
  "response": {
    "direction": {
      "pose_frame": "string",
      "vector": {
        "x": "float64",
        "y": "float64",
        "z": "float64"
      },
      "weight": "float64"
    },
    "gravity": {
      "weight": "float64"
    },
    "match_score": {
      "weight": "float64"
    },
    "preferred_orientation": {
      "weight": "float64"
    },
    "return_code": {
      "message": "string",
      "value": "int16"
    }
  }
}

warmup_template

Loads a template so that it is ready when the first detection with this template is triggered. Without using this service, the first detection with a new template takes longer than the following ones, because the template is then loaded at the first detection.

Details

This service can be called as follows.

PUT http://<host>/api/v2/pipelines/0/nodes/rc_cadmatch/services/warmup_template
PUT http://<host>/api/v1/nodes/rc_cadmatch/services/warmup_template

The definition for the request arguments with corresponding datatypes is:

{
  "args": {
    "template_id": "string"
  }
}

The template_id is the ID of the template to be loaded into the CADMatch module.

The definition for the response with corresponding datatypes is:

{
  "name": "warmup_template",
  "response": {
    "return_code": {
      "message": "string",
      "value": "int16"
    }
  }
}

start

Starts the module. If the command is accepted, the module moves to state RUNNING.

Details

The current_state value in the service response may differ from RUNNING if the state transition is still in process when the service returns.

This service can be called as follows.

PUT http://<host>/api/v2/pipelines/0/nodes/rc_cadmatch/services/start
PUT http://<host>/api/v1/nodes/rc_cadmatch/services/start
This service has no arguments.

The definition for the response with corresponding datatypes is:

{
  "name": "start",
  "response": {
    "accepted": "bool",
    "current_state": "string"
  }
}

stop

Stops the module. If the command is accepted, the module moves to state IDLE.

Details

The current_state value in the service response may differ from IDLE if the state transition is still in process when the service returns.

This service can be called as follows.

PUT http://<host>/api/v2/pipelines/0/nodes/rc_cadmatch/services/stop
PUT http://<host>/api/v1/nodes/rc_cadmatch/services/stop
This service has no arguments.

The definition for the response with corresponding datatypes is:

{
  "name": "stop",
  "response": {
    "accepted": "bool",
    "current_state": "string"
  }
}

reset_defaults

Resets all parameters of the module to its default values, as listed in above table. Also resets preferred orientation and sorting strategies. The reset does not apply to templates.

Details

This service can be called as follows.

PUT http://<host>/api/v2/pipelines/0/nodes/rc_cadmatch/services/reset_defaults
PUT http://<host>/api/v1/nodes/rc_cadmatch/services/reset_defaults
This service has no arguments.

The definition for the response with corresponding datatypes is:

{
  "name": "reset_defaults",
  "response": {
    "return_code": {
      "message": "string",
      "value": "int16"
    }
  }
}

set_region_of_interest (deprecated)

Persistently stores a 3D region of interest on the rc_visard NG.

Details

This service can be called as follows.

This service is not available in API version 2. Use set_region_of_interest in rc_roi_db instead.
PUT http://<host>/api/v1/nodes/rc_cadmatch/services/set_region_of_interest

get_regions_of_interest (deprecated)

Returns the configured 3D regions of interest with the requested region_of_interest_ids.

Details

This service can be called as follows.

This service is not available in API version 2. Use get_regions_of_interest in rc_roi_db instead.
PUT http://<host>/api/v1/nodes/rc_cadmatch/services/get_regions_of_interest

delete_regions_of_interest (deprecated)

Deletes the configured 3D regions of interest with the requested region_of_interest_ids.

Details

This service can be called as follows.

This service is not available in API version 2. Use delete_regions_of_interest in rc_roi_db instead.
PUT http://<host>/api/v1/nodes/rc_cadmatch/services/delete_regions_of_interest

Internal services

The following services for configuring grasps and pose priors can change in future without notice. Setting, retrieving and deleting grasps and pose priors is recommended to be done via the Web GUI.

set_grasp

Persistently stores a grasp for the given object template on the rc_visard NG. All configured grasps are persistent over firmware updates and rollbacks.

Details

This service can be called as follows.

PUT http://<host>/api/v2/pipelines/0/nodes/rc_cadmatch/services/set_grasp
PUT http://<host>/api/v1/nodes/rc_cadmatch/services/set_grasp

Details for the definition of the grasp type are given in Setting of grasp points.

The definition for the request arguments with corresponding datatypes is:

{
  "args": {
    "grasp": {
      "gripper_id": "string",
      "id": "string",
      "pose": {
        "orientation": {
          "w": "float64",
          "x": "float64",
          "y": "float64",
          "z": "float64"
        },
        "position": {
          "x": "float64",
          "y": "float64",
          "z": "float64"
        }
      },
      "priority": "int8",
      "replication": {
        "max_x_deg": "float64",
        "min_x_deg": "float64",
        "origin": {
          "orientation": {
            "w": "float64",
            "x": "float64",
            "y": "float64",
            "z": "float64"
          },
          "position": {
            "x": "float64",
            "y": "float64",
            "z": "float64"
          }
        },
        "step_x_deg": "float64"
      },
      "template_id": "string"
    }
  }
}

The definition for the response with corresponding datatypes is:

{
  "name": "set_grasp",
  "response": {
    "return_code": {
      "message": "string",
      "value": "int16"
    }
  }
}

set_all_grasps

Replaces the list of grasps for the given object template on the rc_visard NG.

Details

This service can be called as follows.

PUT http://<host>/api/v2/pipelines/0/nodes/rc_cadmatch/services/set_all_grasps
PUT http://<host>/api/v1/nodes/rc_cadmatch/services/set_all_grasps

Details for the definition of the grasp type are given in Setting of grasp points.

The definition for the request arguments with corresponding datatypes is:

{
  "args": {
    "grasps": [
      {
        "gripper_id": "string",
        "id": "string",
        "pose": {
          "orientation": {
            "w": "float64",
            "x": "float64",
            "y": "float64",
            "z": "float64"
          },
          "position": {
            "x": "float64",
            "y": "float64",
            "z": "float64"
          }
        },
        "priority": "int8",
        "replication": {
          "max_x_deg": "float64",
          "min_x_deg": "float64",
          "origin": {
            "orientation": {
              "w": "float64",
              "x": "float64",
              "y": "float64",
              "z": "float64"
            },
            "position": {
              "x": "float64",
              "y": "float64",
              "z": "float64"
            }
          },
          "step_x_deg": "float64"
        },
        "template_id": "string"
      }
    ],
    "template_id": "string"
  }
}

The definition for the response with corresponding datatypes is:

{
  "name": "set_all_grasps",
  "response": {
    "return_code": {
      "message": "string",
      "value": "int16"
    }
  }
}

get_grasps

Returns all configured grasps which have the requested grasp_ids and belong to the requested template_ids.

Details

This service can be called as follows.

PUT http://<host>/api/v2/pipelines/0/nodes/rc_cadmatch/services/get_grasps
PUT http://<host>/api/v1/nodes/rc_cadmatch/services/get_grasps

If no grasp_ids are provided, all grasps belonging to the requested template_ids are returned. If no template_ids are provided, all grasps with the requested grasp_ids are returned. If neither IDs are provided, all configured grasps are returned.

The definition for the request arguments with corresponding datatypes is:

{
  "args": {
    "grasp_ids": [
      "string"
    ],
    "template_ids": [
      "string"
    ]
  }
}

The definition for the response with corresponding datatypes is:

{
  "name": "get_grasps",
  "response": {
    "grasps": [
      {
        "gripper_id": "string",
        "id": "string",
        "pose": {
          "orientation": {
            "w": "float64",
            "x": "float64",
            "y": "float64",
            "z": "float64"
          },
          "position": {
            "x": "float64",
            "y": "float64",
            "z": "float64"
          }
        },
        "priority": "int8",
        "replication": {
          "max_x_deg": "float64",
          "min_x_deg": "float64",
          "origin": {
            "orientation": {
              "w": "float64",
              "x": "float64",
              "y": "float64",
              "z": "float64"
            },
            "position": {
              "x": "float64",
              "y": "float64",
              "z": "float64"
            }
          },
          "step_x_deg": "float64"
        },
        "template_id": "string"
      }
    ],
    "return_code": {
      "message": "string",
      "value": "int16"
    }
  }
}

delete_grasps

Deletes all grasps with the requested grasp_ids that belong to the requested template_ids.

Details

This service can be called as follows.

PUT http://<host>/api/v2/pipelines/0/nodes/rc_cadmatch/services/delete_grasps
PUT http://<host>/api/v1/nodes/rc_cadmatch/services/delete_grasps

If no grasp_ids are provided, all grasps belonging to the requested template_ids are deleted. The template_ids list must not be empty.

The definition for the request arguments with corresponding datatypes is:

{
  "args": {
    "grasp_ids": [
      "string"
    ],
    "template_ids": [
      "string"
    ]
  }
}

The definition for the response with corresponding datatypes is:

{
  "name": "delete_grasps",
  "response": {
    "return_code": {
      "message": "string",
      "value": "int16"
    }
  }
}

get_symmetric_grasps

Returns all grasps that are symmetric to the given grasp.

Details

This service can be called as follows.

PUT http://<host>/api/v2/pipelines/0/nodes/rc_cadmatch/services/get_symmetric_grasps
PUT http://<host>/api/v1/nodes/rc_cadmatch/services/get_symmetric_grasps

Details for the definition of the grasp type are given in Setting of grasp points.

The definition for the request arguments with corresponding datatypes is:

{
  "args": {
    "grasp": {
      "pose": {
        "orientation": {
          "w": "float64",
          "x": "float64",
          "y": "float64",
          "z": "float64"
        },
        "position": {
          "x": "float64",
          "y": "float64",
          "z": "float64"
        }
      },
      "replication": {
        "max_x_deg": "float64",
        "min_x_deg": "float64",
        "origin": {
          "orientation": {
            "w": "float64",
            "x": "float64",
            "y": "float64",
            "z": "float64"
          },
          "position": {
            "x": "float64",
            "y": "float64",
            "z": "float64"
          }
        },
        "step_x_deg": "float64"
      },
      "template_id": "string"
    }
  }
}

The first grasp in the returned list is the one that was passed with the service call. If the object template does not have an exact symmetry, only the grasp passed with the service call will be returned. If the object template has a continuous symmetry (e.g. a cylindrical object), only 12 equally spaced sample grasps will be returned.

Details for the definition of the grasp type are given in Setting of grasp points.

The definition for the response with corresponding datatypes is:

{
  "name": "get_symmetric_grasps",
  "response": {
    "grasps": [
      {
        "pose": {
          "orientation": {
            "w": "float64",
            "x": "float64",
            "y": "float64",
            "z": "float64"
          },
          "position": {
            "x": "float64",
            "y": "float64",
            "z": "float64"
          }
        },
        "replication": {
          "max_x_deg": "float64",
          "min_x_deg": "float64",
          "origin": {
            "orientation": {
              "w": "float64",
              "x": "float64",
              "y": "float64",
              "z": "float64"
            },
            "position": {
              "x": "float64",
              "y": "float64",
              "z": "float64"
            }
          },
          "step_x_deg": "float64"
        },
        "template_id": "string"
      }
    ],
    "return_code": {
      "message": "string",
      "value": "int16"
    }
  }
}

set_pose_prior

Persistently stores a pose prior for the given object template on the rc_visard NG. All configured pose priors are persistent over firmware updates and rollbacks.

Details

This service can be called as follows.

PUT http://<host>/api/v2/pipelines/0/nodes/rc_cadmatch/services/set_pose_prior
PUT http://<host>/api/v1/nodes/rc_cadmatch/services/set_pose_prior

Details for the definition of the pose_prior type are given in Setting of pose priors.

The definition for the request arguments with corresponding datatypes is:

{
  "args": {
    "pose_prior": {
      "id": "string",
      "pose": {
        "orientation": {
          "w": "float64",
          "x": "float64",
          "y": "float64",
          "z": "float64"
        },
        "position": {
          "x": "float64",
          "y": "float64",
          "z": "float64"
        }
      },
      "pose_frame": "string",
      "template_id": "string"
    }
  }
}

The definition for the response with corresponding datatypes is:

{
  "name": "set_pose_prior",
  "response": {
    "return_code": {
      "message": "string",
      "value": "int16"
    }
  }
}

set_all_pose_priors

Replaces the list of pose priors for the given object template on the rc_visard NG.

Details

This service can be called as follows.

PUT http://<host>/api/v2/pipelines/0/nodes/rc_cadmatch/services/set_all_pose_priors
PUT http://<host>/api/v1/nodes/rc_cadmatch/services/set_all_pose_priors

Details for the definition of the pose_prior type are given in Setting of pose priors.

The definition for the request arguments with corresponding datatypes is:

{
  "args": {
    "pose_priors": [
      {
        "id": "string",
        "pose": {
          "orientation": {
            "w": "float64",
            "x": "float64",
            "y": "float64",
            "z": "float64"
          },
          "position": {
            "x": "float64",
            "y": "float64",
            "z": "float64"
          }
        },
        "pose_frame": "string",
        "template_id": "string"
      }
    ],
    "template_id": "string"
  }
}

The definition for the response with corresponding datatypes is:

{
  "name": "set_all_pose_priors",
  "response": {
    "return_code": {
      "message": "string",
      "value": "int16"
    }
  }
}

get_pose_priors

Returns all configured pose priors which have the requested pose_prior_ids and belong to the requested template_ids.

Details

This service can be called as follows.

PUT http://<host>/api/v2/pipelines/0/nodes/rc_cadmatch/services/get_pose_priors
PUT http://<host>/api/v1/nodes/rc_cadmatch/services/get_pose_priors

If no pose_prior_ids are provided, all pose priors belonging to the requested template_ids are returned. If no template_ids are provided, all pose priors with the requested pose_prior_ids are returned. If neither IDs are provided, all configured pose priors are returned.

The definition for the request arguments with corresponding datatypes is:

{
  "args": {
    "pose_prior_ids": [
      "string"
    ],
    "template_ids": [
      "string"
    ]
  }
}

The definition for the response with corresponding datatypes is:

{
  "name": "get_pose_priors",
  "response": {
    "pose_priors": [
      {
        "id": "string",
        "pose": {
          "orientation": {
            "w": "float64",
            "x": "float64",
            "y": "float64",
            "z": "float64"
          },
          "position": {
            "x": "float64",
            "y": "float64",
            "z": "float64"
          }
        },
        "pose_frame": "string",
        "template_id": "string"
      }
    ],
    "return_code": {
      "message": "string",
      "value": "int16"
    }
  }
}

delete_pose_priors

Deletes all pose priors with the requested pose_prior_ids that belong to the requested template_ids.

Details

This service can be called as follows.

PUT http://<host>/api/v2/pipelines/0/nodes/rc_cadmatch/services/delete_pose_priors
PUT http://<host>/api/v1/nodes/rc_cadmatch/services/delete_pose_priors

If no pose_prior_ids are provided, all pose priors belonging to the requested template_ids are deleted. The template_ids list must not be empty.

The definition for the request arguments with corresponding datatypes is:

{
  "args": {
    "pose_prior_ids": [
      "string"
    ],
    "template_ids": [
      "string"
    ]
  }
}

The definition for the response with corresponding datatypes is:

{
  "name": "delete_pose_priors",
  "response": {
    "return_code": {
      "message": "string",
      "value": "int16"
    }
  }
}

Return codes

Each service response contains a return_code, which consists of a value plus an optional message. A successful service returns with a return_code value of 0. Negative return_code values indicate that the service failed. Positive return_code values indicate that the service succeeded with additional information. The smaller value is selected in case a service has multiple return_code values, but all messages are appended in the return_code message.

The following table contains a list of common codes:

Return codes of the CADMatch services
Code Description
0 Success
-1 An invalid argument was provided.
-2 An internal error occurred.
-3 An internal timeout occurred.
-4 Data acquisition took longer than allowed.
-8 Not applicable, stereo quality must be at least Medium.
-9 No valid license for the module.
-10 New element could not be added as the maximum storage capacity of load carriers or regions of interest has been exceeded.
-11 Sensor not connected, not supported or not ready.
10 The maximum storage capacity of load carriers or regions of interest has been reached.
11 Existing data was overwritten.
100 The requested load carrier was not detected in the scene.
101 None of the detected grasps is reachable.
102 The detected load carrier is empty.
103 All detected grasps are in collision.
106 The list of returned grasps has been trimmed to the 100 best grasps.
110 Hints for setting up the application, e.g. reducing the distance from the camera, setting a region of interest.
114 No gripper was found for collision checking.
115 Collision checking during retraction was skipped, e.g. because no load carrier or no pre-grasp offset were given.
151 The object template has a continuous symmetry.
152 The objects are outside the given region of interest, outside the load carrier or outside the image.
153 No edges could be detected in the camera image. Check the Edge Sensitivity.
999 Additional hints for application development

Template API

For template upload, download, listing and removal, special REST-API endpoints are provided. Templates can also be uploaded, downloaded and removed via the Web GUI. The templates include the grasp points and pose priors, if grasp points or pose priors have been configured. Up to 50 templates can be stored persistently on the rc_visard NG.