Mobile robot

Abstract

The invention provides a mobile robot. The present invention generally relates to a mobile robot, e.g., a humanoid robot, configured to provide reality capture and metrology-level geometric measurements, e.g., to generally support infrastructure monitoring and / or to support workflows in the metrology field. Aspects of a mobile robot are particularly directed to providing increased accuracy of metrology level devices to overcome defects in mobile reality capture. In another aspect, the mobility advantage provided by the mobile robot is transitioned to the metrology field while maintaining metrology level accuracy.

Description

[0001] The present application is a divisional application of the patent application with the application number 202280102338.0 (international application number: PCT / EP2022 / 084481, filing date: 2022-12-05, invention name: humanoid robot comprising articulated legs with wheels or tracks). TECHNICAL FIELD

[0002] The present invention generally relates to a mobile robot configured to provide reality capture and metrological level geometric measurements, for example to generally support infrastructure monitoring and / or to support workflows in the metrology field. BACKGROUND

[0003] Reality capture is important for monitoring environments, for example to secure restricted or dangerous areas such as industrial plants, construction sites or commercial complexes. Furthermore, the operation of a facility can be supported by providing a quick assessment of the actual state of the environment and using this assessment to provide automated interaction with objects in the environment.

[0004] Reality capture makes use of a combination of various different perception sensing technologies, such as optical and thermal imaging, depth measurement, three-dimensional laser scanning, acoustic sensing, vibration measurement, etc. The referencing and fusion of different data types, for example laser scanner data, camera data and positioning data such as from a global navigation satellite system, is now increasingly standardized.

[0005] Reality capture can be provided by a mobile robot configured to move through an environment and to provide perception data and reference data at the same time, for example, wherein at least the trajectory data of the robot is provided together with the capture of the perception data, so that perception data captured at different positions can be combined into a common coordinate system. Typically, such mobile robots are configured to create a 3D map of a new environment autonomously, for example by means of a simultaneous localization and mapping (SLAM) function.

[0006] As an example, the perception data is analyzed by means of a feature recognition algorithm configured to automatically recognize semantic and / or geometric features captured by the perception data, for example by using shape information provided by virtual object data from CAD models. Such feature recognition, in particular for recognizing geometric primitives, is widely used today.

[0007] In metrology, it is a general goal to determine geometric properties such as coordinates, distances and orientations of one or more target objects. Methods and systems for measuring geometric properties are used in many applications, for example very precise measurements in geodetic applications, measurement problems in the field of building installation, or for controlling or assisting industrial processes.

[0008] Metrology applications often require measurement accuracies that exceed the accuracies provided by real capture devices by orders of magnitude. Typically, measurement devices in the metrology field are heavy and need to be accurately referenced or mounted to a fixed position to provide coordinate measurement data with accuracies better than the millimeter range, often better than the micrometer range. For example, typical metrology measurement devices are laser scanners, structured light scanners, coordinate measuring machines (CMMs) and articulate measurement arms.

[0009] Typically, when using mobile robots for surveillance purposes, the limited measurement accuracy (e.g. compared to metrology devices) is not a problem. For example, the point density and distance measurement accuracy of a 3D laser scanner must be sufficient to be able to recognize a left behind object, but does not need to be able to recognize objects or deformations within the environment in the millimeter or even micrometer range. However, in certain applications, the limited measurement accuracy / ability of a general purpose mobile surveillance robot can cause problems. For example, a general purpose surveillance robot still has problems in recognizing partially open windows and doors.

[0010] On the other hand, it would also be beneficial in the metrology field to have the mobility and increased flexibility provided by mobile real capture robots, for example this would reduce the space required by bulky and heavy metrology devices, increase the flexibility to measure different types of objects and provide more efficient industrial processes. However, the structure of a mobile robot, for example comprising many movable parts and joints, introduces additional degrees of freedom, which make it more difficult to provide accurate measurements. In addition, mobility can cause additional mechanical loads on sensitive metrology sensors, which have to be taken into account. SUMMARY

[0011] It is therefore an object of the present invention to provide a mobile robot that overcomes the deficiencies of prior art robots in the field of real capture, in particular in the field of infrastructure surveillance.

[0012] It is a further object to provide a metrology system that provides increased flexibility and more efficient data acquisition.

[0013] One aspect of the present invention relates to a humanoid robot. The robot comprises a main body, two mechanically actuated leg parts attached to the main body at a lower part of the main body and configured to provide locomotion of the robot on the ground, two mechanically actuated arm parts attached to the main body at an upper part of the main body and configured to move relative to the main body, and a (e.g. mechanically actuated) head part attached to the main body at a top part of the main body.

[0014] Each of the two mechanically actuated leg portions is attached to the body by a hip joint, providing movement of an upper part of the leg portion relative to the body. Each leg portion further comprises a knee joint providing movement of a lower part of the leg portion relative to the upper part of the leg portion, and a wheel at a distal end away from the knee joint for contacting the ground to provide said locomotion.

[0015] The robot is configured to provide said locomotion in a walking mode by stepping motion of the leg portions, and in a driving mode by rolling on the wheels.

[0016] One of the two leg portions, in particular each of the two leg portions (see below), comprises a battery compartment arranged between the knee joint and the hip joint and configured to accommodate a replaceable battery. When the battery is accommodated in the battery compartment, the battery provides electrical energy for driving motion of the robot, e.g. motion of one of the two leg portions (e.g. including motion of the wheel) and / or motion of one of the two arm portions. In particular, the robot is configured to autonomously perform a task involving movement of parts of the robot, and configured such that the replaceable battery provides a substantial part, in particular all, of the electrical energy consumed by the robot during performance of said task. The robot is configured to provide battery replacement for the battery compartment during continuous operation of the robot, e.g. during motion of the two arm portions.

[0017] In one embodiment, the robot comprises a further battery compartment arranged between the knee joint and the hip joint of the other of the two leg portions, i.e. each of the two leg portions comprises a battery compartment. The further battery compartment is configured to accommodate a further replaceable battery. When the further battery is accommodated in the further battery compartment, the further battery provides electrical energy for motion of said robot, in particular for motion of one of said two leg portions (e.g. motion of the leg portion itself and / or motion of the wheel) and / or motion of one of the two arm portions. Each of the battery compartments comprises a circuit breaker configured to be activated for battery replacement of the respective battery compartment and to provide electrical disconnection of the battery located in the respective battery compartment, such that the robot is still supplied with electrical energy from the other battery compartment.

[0018] In another embodiment, the humanoid robot is configured to autonomously remove the battery located in the battery compartment using one of the arm portions and to autonomously place the battery into the battery compartment, in particular wherein the battery compartment comprises a quick release mechanism configured to be activated by the robot.

[0019] In another embodiment, each of the hip joints provides two degrees of rotational freedom of movement relative to the body, and each of the knee joints provides one degree of rotational freedom of movement (e.g. providing a folding movement for the leg portion).

[0020] In another embodiment, each of the knee joints is driven by a respective electric motor positioned distally from the knee joint, and each of the legs comprises a mechanical transmission element, such as a belt or chain, driven by the respective electric motor to provide mechanical actuation of the movement about the respective knee joint.

[0021] In another embodiment, each of the arm portions is attached to the body by a shoulder joint. The shoulder joint provides movement with one degree of rotational freedom relative to the body, in particular movement with two degrees of rotational freedom relative to the body. The arm portion further comprises an elbow joint and a hand joint arrangement. The elbow joint provides movement with one degree of rotational freedom, in particular for folding the arm portion.

[0022] In another embodiment, the hand joint arrangement is configured for performing grasping operations.

[0023] As an example, the robot comprises a further joint, such as a further joint arranged on the arm portion to provide a degree of rotational freedom of the arm portion about an arm portion axis. For example, the further joint is arranged between the elbow joint and the shoulder joint or between the elbow joint and the hand joint. For example, the arm portion axis is coaxial with an imaginary line intersecting the elbow joint and the hand joint, or coaxial with an imaginary line intersecting the shoulder joint and the elbow joint.

[0024] In another embodiment, each of the shoulder joint, the elbow joint and the hand joint comprises a robot drive module for driving rotational joint movement. The robot drive module comprises a rotary drive comprising a motor circuit board, a stator and a rotor. The rotor is configured to be controlled by the motor circuit board to rotate relative to the stator about a rotation axis. The robot drive module further comprises a gearbox configured to convert rotational motion of the rotor about the rotation axis to rotational motion of a gearbox output member about the rotation axis according to a defined gear ratio. The motor circuit board and the stator are arranged axially relative to the rotation axis on one side of the gearbox, denoted as a gearbox input side. The gearbox output member engages the gearbox from the other side of the gearbox, denoted as a gearbox output side. The robot drive module further comprises a connection extending from the gearbox output side to the gearbox input side and configured to pick up rotation of the gearbox output member in a rigid manner, thereby providing rotation of the connection in line with, e.g. identical to, the rotation of the gearbox output member. The robot drive module further comprises a rotary encoder configured to detect rotation of the gearbox output member about the rotation axis. The rotary encoder is arranged on the gearbox input side and configured for measuring rotation of the connection about the rotation axis.

[0025] In another embodiment, at least one of the arm portions comprises a probing sensor device, in particular at least one of the arm portions comprises a probing sensor device at a distal end distal from an attachment point of the arm portion, in particular the probing sensor device comprises an optical sensor configured to provide optical sensor data and / or a haptics sensor configured to provide haptics scan data. In the sense of the present invention, optical comprises the whole electromagnetic spectrum, e.g. from ultraviolet to infrared wavelength ranges and also beyond. By way of example, an optical sensor in the sense of the present invention is embodied as an imaging sensor, e.g. a 2D or 3D imaging camera, or a laser-based scanning sensor, e.g. a scanning sensor for detecting reflected light from a cooperating target and / or for detecting diffuse scattered light. In particular, a probing sensor in the sense of the present invention relates to a sensor configured to provide coordinate measurements by approaching an object to be measured and determining a relative distance and / or a relative orientation of the sensor to a surface of the measured object. For example, by single point measurements to probe a relative distance of a measurement point and thus to probe a coordinate of the measurement point. Alternatively, the probing sensor is moved over a surface to be measured and provides continuous sensor data to evaluate a relative distance of a trajectory along which the sensor is moved to the object surface. In particular, a probing sensor in the sense of the present invention is embodied as a metrology grade sensor, e.g. a sensor with sub-millimeter distance resolution.

[0026] In another embodiment, the robot comprises a scanning laser radar unit configured to provide a scanning movement of a laser measuring beam relative to two rotational axes during movement of the robot on the ground and based on the scanning movement, to generate light detection and ranging data for generating a three-dimensional point cloud. The laser radar unit is preferably located in an upper portion of the robot, in particular in the main body and / or the head.

[0027] In another embodiment, the robot comprises a ToF unit. The ToF unit comprises an arrangement of time-of-flight sensors and is configured to provide 3D imaging data of the environment during movement of the robot on the ground to generate a 3D model of the environment. The ToF unit is preferably located in an upper portion of the robot, in particular in the main body and / or the head.

[0028] For example, the head comprises a top side opposite an attachment point of the head to the main body, a front side adjacent to the top side and a rear side opposite the front side and adjacent to the top side and two opposite lateral sides adjacent to the top side and the front side and the rear side, respectively.

[0029] In another embodiment, each of the lateral sides comprises an imaging camera, such that the two imaging cameras are arranged opposite to each other and provide opposite fields of view, in particular wherein each of the two cameras provides a complete spherical 360° field of view.

[0030] In another implementation, one of the two rotational axes of the laser radar unit extends through the front side and the rear side of the head, and the laser radar unit is configured to provide a front view field of view around the one of the two rotational axes, wherein the front side faces in a direction of travel during forward travel.

[0031] In another implementation, the ToF unit is configured to provide a front view field of view, wherein the front side faces in a direction of travel during travel.

[0032] In another implementation, the body and / or the head comprises a structured light scanner configured to provide a 3D scan of the environment during movement of the robot on the ground to generate a 3D model of the environment.

[0033] In another implementation, the robot comprises a simultaneous localization and mapping unit, SLAM unit. The SLAM unit is configured to perform a simultaneous localization and mapping process (SLAM process). The SLAM process comprises receiving perception data providing a representation of the surroundings of the robot at a current position, generating a map of the environment using the perception data, and localizing the robot within the map of the environment.

[0034] In another implementation, the robot is further configured to: a.) access information about a location of a charging and / or battery exchange station within the map of the environment, in particular information about a charging location and / or availability of an exchange battery in individual ones of the stations, b.) localize itself relative to a location of at least one charging and / or battery exchange station within the map of the environment and provide an assessment of a reachability of the at least one charging and / or battery exchange station based on a state of charge of an exchangeable battery, and c.) trigger a movement to the station based on the reachability of the at least one charging and / or battery exchange station, in particular wherein the reachability is below a range threshold.

[0035] In another implementation, the robot comprises a charging element, in particular a wired connector. The charging element is configured to provide electrical energy during the battery exchange, e.g., wherein the robot is configured to autonomously connect the charging element to a charging location during the battery exchange.

[0036] In another implementation, the robot comprises a backup battery configured to provide electrical energy during the battery exchange. The backup battery can be another battery in the battery compartment that is not exchanged during the battery exchange, e.g., for implementations wherein each leg comprises an exchangeable battery. The backup battery can also be an emergency power supply configured of another battery during the battery exchange.

[0037] In another implementation, each of the leg portions comprises a plurality of wheels for contacting the ground to provide said locomotion.

[0038] In another implementation, each of the leg portions comprises a three-wheeled device. The three-wheeled device comprises: a.) three wheels attached to a wheel carrier configured to support the three wheels in a circular pattern, b.) a main shaft (connecting the wheel carrier to the rest of the leg portion) providing a rotational degree of freedom of the wheel carrier, and c.) three wheel shafts different from the main shaft, wherein the wheel shafts connect the wheels to the wheel carrier, wherein each of the wheel shafts provides a wheel rotational degree of freedom independent of the rotational degree of freedom of the main shaft and the wheel rotational degrees of freedom of the other two wheel shafts.

[0039] In another implementation, the robot is configured to provide self-balancing movement control for the movement of the arm portions during locomotion. The self-balancing movement supports the robot to balance in a defined upright position. The robot comprises inertial sensors (in particular, tilt sensors) and gyroscopic sensors and a control algorithm configured to automatically adjust the relative attitude of the two mechanically actuated arm portions based on the inertial sensors and the gyroscopic sensors to perform self-balancing movement.

[0040] Separately or in combination with other aspects of the invention, another aspect of the aspects of the invention relates to a humanoid robot comprising a main body and mechanically actuated leg portions. The mechanically actuated leg portions are attached to the main body at a lower portion of the main body and are configured to provide locomotion of the robot on the ground. The leg portions comprise an upper portion and a lower portion, which are connected to each other via a joint and can swivel relative to each other about the joint.

[0041] The lower portion comprises a track drive. The track drive comprises a track running over a lower pulley and an upper pulley, thereby providing a running surface between the lower pulley and the upper pulley. The lower pulley is arranged on the lower portion at a distal end distal from the joint, and the upper pulley is arranged closer to the joint than the lower pulley.

[0042] The robot is configured to provide locomotion by track travel in a stand mode and a full-track mode. The stand mode provides a surface contact of a running surface having a smaller area dimension than a surface contact of a running surface provided by the full-track mode. The stand mode is achieved by the robot automatically arranging the upper part relative to the lower part such that during locomotion the upper pulley is elevated to an elevated position further away from the ground than the position of the lower pulley. In particular, in the stand mode the robot supports itself by self-balancing locomotion to maintain the elevated position of the upper pulley, more particularly wherein in the stand mode the robot supports itself by standing on a curved portion of the track only, which is curved by a circumferential area of the lower pulley.

[0043] In another embodiment, in the stand mode the robot is configured to support itself by self-balancing locomotion to maintain the elevated position of the upper pulley, in particular for which the robot comprises inertial sensors and gyroscopic sensors and a control algorithm configured to automatically control the track drive (e.g. lower pulley) based on the inertial sensors and the gyroscopic sensors such that the robot balances in a defined upright position associated with the elevated position of the upper pulley.

[0044] In another embodiment, the robot is configured to maintain an orientation of the main body relative to a defined attitude during the locomotion, the defined attitude being relative to a direction of gravity, e.g. wherein at the same time an orientation of the running surface relative to the ground (i.e. the surface contact) is maintained.

[0045] In another embodiment, the robot comprises a ground characterization sensor configured to provide ground quality information, in particular a parameter providing wheelspin information. The ground quality information is based on e.g. a hardness and / or roughness and / or slope of the ground. The robot is configured to automatically set different surface contacts of the running surface, i.e. different area dimensions of the running surface contacting the ground, based on the ground quality information, in particular by providing different elevated positions of the upper pulley above the ground relative to the lower pulley and self-balancing in each of the different elevated positions.

[0046] In another embodiment, the ground characterization sensor comprises a set of inertial sensors. The set of inertial sensors is configured to provide a.) track slip information, wherein track slip represents a relative motion of a portion of the track engaging the ground with the ground, and b.) impact information, wherein impact represents an acceleration above a threshold value, in particular in a vertical direction.

[0047] In another embodiment, the ground property sensor comprises an optical sensor device, in particular comprising a stereo camera and / or a time-of-flight camera and / or a lidar sensor, configured to provide a view of the ground in the direction of motion of the robot, wherein the robot is configured to a.) use the optical sensor device to provide a prediction of ground quality information about a subsequent path to be traveled by the robot, and b.) automatically set one of the different surface contacts based on the prediction.

[0048] In another embodiment, the robot comprises a further mechanically actuated leg attached to the main body at a lower portion of the main body. The further mechanically actuated leg comprises an upper portion and a lower portion connected to each other via a joint and pivotable relative to each other about the joint. The lower portion comprises a further track drive comprising a track running over a lower pulley and an upper pulley, thereby providing a running surface between the lower pulley and the upper pulley. The lower pulley is arranged on the lower portion at a distal end distal from the joint, and the upper pulley is arranged closer to the joint than the lower pulley. The robot is configured to adapt a surface contact of the running surface of the further track drive by arranging the lower portion of the further leg relative to the upper portion of the further leg such that, during travel, the upper pulley of the further track drive is elevated to an elevated position further away from the ground than a position of the lower pulley of the further track drive, in particular wherein the robot is configured to provide travel in a walking mode by a stepwise motion of the leg and the further leg.

[0049] In another embodiment, the robot comprises a further mechanically actuated leg attached to the main body at a lower portion of the main body. The further mechanically actuated leg comprises an upper portion and a lower portion connected to each other via a joint and pivotable relative to each other about the joint. The lower portion comprises a further track drive comprising a track running over a lower pulley and an upper pulley, thereby providing a running surface between the lower pulley and the upper pulley. The lower pulley is arranged on the lower portion at a distal end distal from the joint, and the upper pulley is arranged closer to the joint than the lower pulley. The robot is configured to adapt a surface contact of the running surface of the further track drive by arranging the lower portion of the further leg relative to the upper portion of the further leg such that, during travel, the upper pulley of the further track drive is elevated to an elevated position further away from the ground than a position of the lower pulley of the further track drive, in particular wherein the robot is configured to provide travel in a walking mode by a stepwise motion of the leg and the further leg.

[0050] The SLAM unit is configured to perform a simultaneous localization and mapping process (SLAM process). The SLAM process comprises receiving perception data providing a representation of a surrounding environment of the robot at a current position, generating a map of the environment using the perception data, and localizing the robot within the map of the environment.

[0051] The laser radar device is configured to generate laser radar data to provide a coordinative scan of the environment with respect to the laser radar device. The robot is configured to generate a 3D model of the environment based on the laser radar data and comprises a classification algorithm. The classification algorithm is configured to: a.) automatically identify a movable barrier object, e.g. a door or window, controlling an entry and exit between different parts of the environment within the 3D model of the environment, b.) assign a geometric test parameter and an associated test criterion to the barrier object, wherein the geometric test parameter provides geometric information, wherein the test criterion provides an assessment of a blocking state of the barrier object with respect to the entry and exit between different parts of the environment depending on the geometric test parameter, and c.) determine the blocking state of the barrier object by determining a value of the geometric test parameter and assessing this value based on the test criterion. The robot is further configured to automatically intervene the barrier object to change the blocking state and / or to provide a data communication with an external device to forward the blocking state.

[0052] In another embodiment, the geometric test parameter and the associated test criterion are an angle information and / or a distance information of a part of the barrier object with respect to a surrounding environment of the barrier object, respectively.

[0053] In another embodiment, the classification algorithm is configured to identify a control component on the barrier object exclusively foreseen for opening and closing the barrier object. The robot is configured to automatically intervene the control component to change the blocking state, in particular wherein the mobile robot comprises a mechanically actuated gripping component.

[0054] In another embodiment, the classification algorithm is configured to assign a geometric test parameter and an associated test criterion to the control component. The geometric test parameter and the associated test criterion of the control component are used to associate a geometric pose of the control component with the blocking state.

[0055] In another embodiment, the robot is configured to: a.) access a 3D reference model, wherein the 3D reference model represents an environment with the barrier object in a reference state, in particular wherein the 3D reference model is stored on the robot, or wherein the robot is configured to access an external data storage unit comprising the 3D reference model, b.) determine an alarm state based on a difference analysis identifying a difference between the 3D reference model and the 3D model of the environment, the difference being generated based on laser radar data, in particular a difference with respect to a geometric pose of the movable barrier object or the control component, and c.) include the alarm state in a data communication with the external device

[0056] In another embodiment, the mobile robot is configured to: a.) access a digital reference map of an environment comprising a set of movable obstacle objects and to localize itself within said reference map, b.) derive an inspection path based on said reference map, wherein said inspection path provides a passage through individual movable obstacle objects of said set of obstacle objects, and c.) automatically follow said inspection path and determine a blocking state of individual obstacle objects of said set of obstacle objects.

[0057] In another embodiment, the reference map comprises a plurality of movable obstacle objects and said robot comprises a user interface providing a selection of said set of obstacle objects from said plurality of obstacle objects.

[0058] In another embodiment, the robot is configured to access a nominal state of said obstacle objects and to automatically intervene said obstacle objects to establish said nominal state in case the determined blocking state deviates from said nominal state.

[0059] In another embodiment, the robot is configured to: a.) provide a time stamp of a generation of said LIDAR data under said 3D model of said environment to said 3D model of said environment, b.) provide a target state of said blocking state of said obstacle objects based on said time stamp, in particular based on a defined schedule providing different target states depending on a time of day, e.g. based on a day / night schedule or based on an evaluation of said 3D model of said environment generated based on older LIDAR data, and b.) in case the determined blocking state deviates from said target state, automatically intervene said obstacle objects to establish said target state and / or provide a report to said external device based on an evaluation of a conformity of the determined blocking state with said target state.

[0060] In another embodiment, the robot is configured to intervene an obstacle object based on at least one of: a.) temperature data of said environment, e.g. by closing a window in case the temperature is below a threshold, b.) humidity data of said environment; and c.) weather forecast data, in particular based on a weather alert provided by said weather forecast data.

[0061] Separately or in combination with other aspects of the application, another aspect of the aspects of the application relates to a mobile robot, in particular embodied as a humanoid robot. The mobile robot comprises a body, a mechanically actuated articulated arm portion, a limb position determination device, and a locomotion unit configured to provide locomotion of the robot on the ground.

[0062] A mechanically actuated articulated arm portion is attached to a body and configured to move relative to the body. The articulated arm portion is attached to the body by a joint, e.g. a pivot joint or a revolute joint, and comprises two arm segments connected to each other by another joint. Each of the arm portion and a robot component different from the arm portion, referred to as another sensor accommodating component, e.g. the body, a head attached to the body at an upper portion of the body or another arm portion, comprises a probing sensor configured to provide distance probing of a surface of an object to be measured by the robot. For example, the probing sensor of the arm portion is of the same or a different type as the probing sensor of the other sensor accommodating component. In the sense of the invention, probing sensor and distance probing means a metrological level of measurement, e.g. with sub-millimeter accuracy, to provide coordinate measurement by approaching the object to be measured and determining the relative distance and / or relative orientation of the sensor to the surface of the measured object. For example, the relative distance is probed by a single point measurement and thus the coordinates of the measurement point are probed. Alternatively, the probing sensor is moved over the surface to be measured and provides continuous sensor data to evaluate the relative distance of the trajectory along which the sensor is moved to the surface of the object.

[0063] The arm segment position determination device comprises a plurality of angle encoders to measure the angular position of the joint and the other joint of the articulated arm portion.

[0064] The robot comprises a reference function comprising: a.) scanning the arm portion by means of the probing sensor of the other sensor accommodating component, and / or b.) scanning a robot portion, e.g. a portion of the body, a portion of the head or a portion of the other arm portion, by means of the probing sensor of the arm portion, and c.) combining the data of the arm segment position determination device with the results of the scanning of the arm portion and / or the scanning of the robot portion to determine reference data for the position determination of the probing sensor of the arm portion, in particular reference data for the compensation of positioning errors of the arm segment position determination device.

[0065] For example, the scanning of the arm portion and / or the scanning of the robot portion is used to calibrate the arm segment position determination device to improve the position determination of the articulation of the arm portion by using the arm segment position determination device. In other words, the control and measurement of the articulation of the arm portion in space provided by the arm segment position determination device is improved. Alternatively or in addition, the position determination of the probing sensor of the arm portion, e.g. such as shown in the arrangement in the hand of the arm portion in Figs. 1 and 2, is performed based on the data provided by the arm segment position determination device and the scanning of the arm portion and / or the scanning of the robot portion. For example, different arm portion positions and robot states are measured by the arm segment position determination device and the probing sensor of the other sensor accommodating component to minimize the positioning errors of the probing sensor of the arm portion. Figure 6 and Figure 7 For example, different arm portion positions and robot states are measured by the arm segment position determination device and the probing sensor of the other sensor accommodating component to minimize the positioning errors of the probing sensor of the arm portion.

[0066] In another embodiment, the probe sensor of the further sensor accommodating component is embodied as a tracker, in particular a laser tracker or a camera-based tracker, the tracker being configured to track a target area, in particular being configured to provide a tracking signal to a cooperative target of the laser tracker. As an example, the cooperative target can be a passive reflective unit with defined reflective properties, such as a steel ball of known size or a retro-reflective unit such as a cube corner, wherein at least part of a laser beam emitted by the laser tracker is reflected back to the laser tracker, e.g. in parallel. For example, the robot comprises one or more trackers, e.g. embodied as an etalon tracer or other interferometric unit, which can follow the “finger probe point” to obtain spatial accuracy of the “finger probe point” in microns. In case of multiple trackers, the tracker positions are stabilized relative to each other during the reference function.

[0067] In a further embodiment, the reference function comprises the robot autonomously moving the arm portion and performing a scan of the robot portion using the probe sensor of the arm portion, in particular wherein the probe sensor of the arm portion is embodied as an optical sensor or a haptic sensor and the scan of the robot portion comprises approaching and measuring different positions on the robot portion by the probe sensor of the arm portion. As an example, the probe sensor in the hand of the robot (e.g. as shown close to the finger) is used to touch / probe a part of the robot itself (not at the same time as the robot is making measurements to measure the object to be measured), wherein the robot touches itself with different articulations (i.e. positions of the joint and the further joint). These touch / probe measurements are then used to calibrate or re-calibrate the arm segment position determination means or the position of the probe sensor of the arm portion relative to the body. For example, such measurements are used to compensate for effects due to the environment and aging of the robot. Figure 6 or Figure 7 As an example, the probe sensor in the hand of the robot (e.g. as shown close to the finger) is used to touch / probe a part of the robot itself (not at the same time as the robot is making measurements to measure the object to be measured), wherein the robot touches itself with different articulations (i.e. positions of the joint and the further joint). These touch / probe measurements are then used to calibrate or re-calibrate the arm segment position determination means or the position of the probe sensor of the arm portion relative to the body. For example, such measurements are used to compensate for effects due to the environment and aging of the robot.

[0068] In another embodiment, the robot is configured to provide a plurality of settable measurement arrangements, the plurality of settable measurement arrangements differing from each other by providing different positioning of the joint and the further joint. The reference function comprises: a.) setting one of the settable measurement arrangements and controlling the travel unit to move the robot relative to the object to be measured such that the robot assumes a calibration position, in which the probe sensors have an unobstructed view of each other, and b.) performing a scan of the arm portion and / or a scan of the robot portion in the calibration position

[0069] In another embodiment, the reference function comprises: a.) scanning, using the probe sensor of the further sensor hosting component, an area of the arm portion spaced apart from the probe sensor of the arm portion to provide 3D information of the area, b.) automatically associating the area with a relative position parameter providing a relative position of the probe sensor of the arm portion with respect to the area by taking into account the 3D information, and using the relative position parameter for determining the reference data.

[0070] In another embodiment, the arm portion comprises a pattern, in particular a visual pattern. The robot is configured to: a.) automatically identify a pattern portion within a measurement of the pattern by the probe sensor of the further sensor hosting component, and b.) automatically associate the pattern portion to the position parameter, wherein the position parameter provides a relative position of the pattern portion with respect to the probe sensor of the arm portion.

[0071] In another embodiment, the robot is configured to access a 3D model of the arm portion and to automatically identify an arm portion feature within a measurement by the probe sensor of the further sensor hosting component. The reference function comprises: a.) automatically associating the arm portion feature with an arm portion feature within the 3D model of the arm portion, b.) based thereon, associating the arm portion feature with a further relative position parameter providing a relative position of the arm portion feature with respect to the probe sensor of the arm portion, and c.) using the further relative position parameter for determining the reference data.

[0072] In another embodiment, the distance probing is provided by laser-based scanning and / or camera-based scanning.

[0073] In a further embodiment, the robot comprises a further mechanically actuated articulated arm portion attached to the main body. The further arm portion is different from the further sensor hosting component. The further arm portion is attached to the main body by a joint and comprises two arm segments connected to each other by a further joint. The arm portion and the further arm portion are configured to move independently from each other with respect to the main body. The arm segment position determining device comprises a plurality of angle encoders to measure angular positions of the joint and the further joint of the further arm portion. The further arm portion comprises a probe sensor configured to provide distance probing of a surface of an object to be measured by the robot. Thus, the robot comprises at least three probe sensors, namely one probe sensor comprised by each arm portion and one probe sensor comprised by a further sensor hosting component, e.g. the main body, a leg portion of the robot or a head portion of the robot. The reference function comprises: a.) mutually scanning the arm portion and the further sensor hosting component by the probe sensors, and b.) using results of the mutual scanning for determining the reference data.

[0074] In another aspect, the aspects of the application relate to a mobile robot, in particular embodied as a humanoid robot. The mobile robot comprises a locomotion unit configured to provide locomotion of the robot over ground, a mechanically actuated multi-joint articulation system, a position determination device, and a coupling interface.

[0075] The mechanically actuated multi-joint articulation system comprises a plurality of joints and is configured to provide movement of an interaction component relative to a reference point on the robot. The interaction component comprises a probing sensor. The probing sensor is configured to provide optical and / or tactile distance probing of a surface of an object to be measured by the robot.

[0076] The coupling interface is configured to provide docking of a docking point of the robot to a coupling counterpart device, e.g. a coupling counterpart device arranged at a measurement platform configured to support an object to be measured. The interaction component is connected to the docking point via a subset of the plurality of joints. The degrees of freedom of movement of the interaction component relative to the docking point depend on joint positions of the subset of the plurality of joints and are independent of joint positions of the remainder of the plurality of joints. The robot is configured to provide a 3D scan by moving the subset of the plurality of joints and by the probing sensor taking measurements.

[0077] The position determination device is configured to provide determination of angular positions of the plurality of joints. The position determination device is configured to provide increased accuracy of determination of angular positions for the subset of the plurality of joints compared to the remainder of the plurality of joints, in particular in a docked state of the robot.

[0078] In another embodiment, the docking is a releasable mechanical docking, and in the docked state the interaction component is mechanically connected to the coupling counterpart device via the subset of the plurality of joints and the coupling interface.

[0079] In another embodiment, the coupling interface is configured to provide the docking as a rigid docking with the coupling counterpart device, in particular by three-point support between the coupling interface and the coupling counterpart device.

[0080] In another embodiment, the coupling interface is configured to provide that all six degrees of freedom in space are fixed when the coupling interface is docked to the coupling counterpart device.

[0081] In another embodiment, the coupling interface comprises a coupling sensor unit, in particular comprising a video camera and / or a capacitive sensor with a measurement range of less than 1 cm, the coupling sensor unit being configured to continuously determine a 6DoF change of position relative to the coupling counterpart device.

[0082] In another embodiment, each joint of the subset of the plurality of joints comprises a robotic drive module for driving rotational joint movement. The robotic drive module comprises a rotary drive comprising a motor circuit board, a stator and a rotor. The rotor is configured to be controlled by the motor circuit board to rotate relative to the stator about a rotary axis. The robotic drive module further comprises a gearbox configured to convert rotational movement of the rotor about a rotary axis into rotational movement of a gearbox output member about the rotary axis according to a defined gear ratio. The motor circuit board and the stator are arranged axially relative to the rotary axis on one side of the gearbox, denoted as a gearbox input side. The gearbox output member engages the gearbox from another side of the gearbox, denoted as a gearbox output side. The robotic drive module further comprises a connection extending from the gearbox output side to the gearbox input side and configured to pick up rotation of the gearbox output member in a rigid manner, thereby providing rotation of the connection in line with, in particular identical to, the rotation of the gearbox output member. The robotic drive module further comprises a rotary encoder configured to detect rotation of the gearbox output member about the rotary axis. The rotary encoder is arranged on the gearbox input side and configured for measuring rotation of the connection about the rotary axis.

[0083] In another embodiment, the robot comprises a laser tracker. The laser tracker is configured to be arranged in a fixed positional relationship with the docking point. For example, the laser tracker is rigidly mounted on the robot or arranged on a movable component of the robot that provides an end stop having a fixed positional relationship with the docking point. The laser tracker is configured to provide automated laser-based tracking of a portion of the robot moved by at least one joint of the subset of the plurality of joints and to determine 3D positional data of the portion of the robot.

[0084] In another embodiment, the robot comprises a camera arrangement. The camera arrangement is configured to be arranged in a fixed positional relationship with the docking point. For example, the camera is rigidly mounted on the robot or arranged on a movable component of the robot that provides an end stop having a fixed positional relationship with the docking point. The camera arrangement is configured to provide automated image-based tracking of another portion of the robot moved by at least one joint of the subset of the plurality of joints and to determine 3D positional data of the other portion of the robot, in particular wherein the other portion comprises a visual pattern for assisting the image-based tracking.

[0085] In another embodiment, the robot comprises a further interaction component. The further interaction component is configured to grab an object and to move by moving a plurality of joints of said multi-jointed articulation. Said further interaction component is connected to said docking point via a further subset of said plurality of joints. The degree of freedom of movement of said further interaction component relative to said docking point depends on the joint positions of said further subset of said plurality of joints and is independent of the joint positions of said subset of said plurality of joints and of a remainder of said plurality of joints. The position determining device is configured to provide an increased accuracy of determination of angular positions for said further subset of said plurality of joints compared to said remainder of said plurality of joints, in particular in a docked state of the robot. In particular, the accuracy of determination of positions of the further subset of the plurality of joints is similar to the above-described subset of the plurality of joints. The robot is configured to provide said 3D scan by grabbing said object to be measured by said further interaction component and moving said further subset of said plurality of joints. The further interaction component can comprise a gripping or alternative fixation element configured to provide a rigid pose relationship between the further interaction component and the object to be measured.

[0086] In another embodiment, the robot is configured to derive pose information of said coupling counterpart relative to a mounting platform carrying said coupling interface. The coupling counterpart is arranged on said mounting platform and is configured to interact with said coupling interface to provide said docking, in particular wherein said robot is configured to access said pose information by electronic communication through said docking point.

[0087] In another embodiment, the robot comprises an optical perception sensor. The optical perception sensor is configured to a.) use said perception sensor for pattern recognition and pattern analysis of a pattern arranged on said coupling counterpart and / or said mounting platform, and b.) to associate said pattern with a pose parameter providing said pose information, in particular by accessing a database comprising different patterns and associated pose parameters for individual ones of said different patterns.

[0088] In another implementation, a coupling counterpart arrangement is arranged at a measurement platform configured to support the object to be measured. The measurement platform comprises a first coupling counterpart and a second coupling counterpart, wherein the first coupling counterpart and the second coupling counterpart are arranged spaced apart from each other. Each of the first coupling counterpart and the second coupling counterpart is configured to interact with the coupling interface to provide a docking of the docking point of the robot to the measurement platform. The robot is configured to: a.) derive first pose information of the first coupling counterpart relative to the measurement platform and derive second pose information of the second coupling counterpart relative to the measurement platform; and b.) provide 3D scan data for the object to be measured by: i) docking to the first coupling counterpart and providing first distance probe data of a first docking position relative to the first coupling counterpart by distance probing with the probe sensor, ii) docking to the second coupling counterpart and providing second distance probe data of a second docking position relative to the second coupling counterpart by distance probing with the probe sensor, and iii) merging the first distance probe data and the second distance probe data using the first pose information and the second pose information.

[0089] In another implementation, the robot is configured to perform a docking pose adjustment measurement. The docking pose adjustment measurement comprises: a.) selecting a set of basic features, wherein i) the first distance probe data comprises a pose of a feature of the set of basic features relative to the first coupling counterpart, and ii) the probe sensor has access to the feature of the set of basic features when the robot is docked to the second coupling counterpart, b.) comparing poses of features of the set of basic features in the first distance probe data and the second distance probe data, and c.) using the comparison to merge the first distance probe data and the second distance probe data.

[0090] Separately or in combination with other aspects of the application, another aspect of the aspects of the application relates to a system for providing maintenance information for a plurality of machines in a factory environment. The system comprises a mobile robot (e.g. a humanoid robot), a microphone arrangement, an event detector, and a positioning unit.

[0091] The microphone arrangement comprises a plurality of microphones and is configured to acquire sound produced within the factory environment. The system is configured to provide sound data based on the acquired sound using the microphone arrangement.

[0092] The event detector is configured to analyze the sound data and, based on the sound data, automatically identify a singled-out noise associated with one of the plurality of machines in the noise data if the singled-out noise deviates from a defined norm. The localization unit is configured to determine a location of the singled-out noise within the factory environment based on sound triangulation.

[0093] The mobile robot comprises a mechanically actuated arm portion, a travel unit, and an optical perception unit. The mechanically actuated arm portion is configured to move relative to a reference point on the robot, the travel unit is configured to provide travel of the robot over a ground surface, and the optical perception unit is configured to provide perception of a portion of an environment surrounding the robot. The robot is configured to automatically move to different locations within the factory environment. The robot further comprises a vibration sensor arranged on the mechanically actuated arm portion. The robot is configured to: a.) identify, by utilizing the optical perception unit, one of the machines located at the location of the singled-out noise, b.) assign a vibration measurement point on the machine to the one of the machines, and c.) bring a probing member of the vibration sensor into contact with the vibration measurement point to generate vibration data of the one of the machines. The system is configured to determine a maintenance status of the one of the machines based on the vibration data.

[0094] In another embodiment, the vibration data is generated by direct motion coupling.

[0095] In another embodiment, the system is configured to determine the maintenance status by comparing the vibration data to comparison data associated with the vibration measurement point.

[0096] In another embodiment, the comparison data is derived from a history of vibration data associated with a defined task performed by the one of the plurality of machines, in particular from a history of vibration data associated with a defined task performed by the one of the plurality of machines or a machine of the plurality of machines having a same machine type as the one of the plurality of machines.

[0097] In another embodiment, the system is configured to train a vibration model using a machine learning algorithm, the vibration model providing an indicator for different maintenance statuses based on a history of sound data of the one of the plurality of machines and a history of the vibration data, in particular by including operator feedback indicative of an abnormal status and a normal status.

[0098] In another embodiment, the system is configured to determine the maintenance status by performing a correlation analysis to determine a correlation of the isolated noise and the vibration data, in particular by determining the maintenance status using a matched filter with a matching template selected from a library of templates for different machine states, more particularly wherein the system is configured to update the library of templates upon detection of a new machine state, e.g. based on a machine learning model and / or based on operator feedback.

[0099] In another embodiment, the system further comprises a database. The robot is configured to access the database, and the database provides an association of one or more vibration measurement points with each of the plurality of machines, in particular by a look-up table.

[0100] In another embodiment, at least a portion of the plurality of microphones is specifically foreseen to be installed in the factory environment.

[0101] In another embodiment, at least a portion of the plurality of microphones is arranged on the robot, and the robot is configured to patrol the factory environment in a defined patrol pattern, during which the robot acquires sound from the factory environment, in particular wherein the robot comprises the event detector.

[0102] In another embodiment, the robot comprises a unidirectional microphone, in particular a unidirectional microphone with a sensitivity below 40 dBA, and the localization unit is configured to determine the location of the isolated noise based on an orientation and a sound level of the unidirectional microphone.

[0103] In another embodiment, the robot is configured to: a.) access or derive a digital model of the factory environment, in particular based on virtual perception data, such that the digital model comprises locations of the plurality of machines, b.) access or derive a patrol pattern comprising a plurality of sound acquisition points, wherein the sound acquisition points are provided such that sound produced by each of the plurality of machines is acquired at at least one sound acquisition point of the patrol pattern, and c.) execute the patrol pattern based on the digital model of the factory environment, such that at each of the sound acquisition points, sound produced within the factory environment is acquired for the event detector and the localization unit.

[0104] Separately or in combination with other aspects of the application, another aspect of the aspects of the application relates to a system comprising a computing unit and a mobile robot, in particular a humanoid robot.

[0105] The robot comprises a mechanically actuated arm portion configured to move relative to a reference point on the robot, a locomotion unit configured to provide locomotion of the robot over a ground surface, an optical perception unit configured to provide perception data for optical perception of a portion of a surrounding environment of the robot, and a power cable.

[0106] The SLAM unit is configured to perform a Simultaneous Localization and Mapping process, i.e. a SLAM process. The SLAM process comprises receiving the perception data providing a representation of the surrounding environment of the robot at a current location, generating a map of an environment using the perception data, and localizing the robot within the map of the environment.

[0107] The mechanically actuated arm portion is configured to autonomously insert the power cable into a power outlet by means of the optical perception unit.

[0108] The computing unit is configured to a.) automatically identify power outlets within the perception data that are compatible with the power cable, b.) determine locations of the power outlets within the map of the environment and store the locations of the power outlets in the map of the environment, and c.) continuously track a relative position of the robot with respect to the locations of the power outlets stored in the map of the environment and schedule the robot to one of the locations of the power outlets stored in the map of the environment depending on a remaining power level of the robot.

[0109] In another embodiment, the robot is configured to charge at a charging station. The system is configured to schedule the robot to the one of the locations of the power outlets by evaluating an energy consumption to reach the charging station and an energy consumption to reach the one of the locations of the power outlets, in particular based on a threshold criterion of the remaining power level of the robot.

[0110] In a further embodiment, the robot comprises a plurality of power cables, each of the plurality of power cables being configured to be compatible with a different power outlet, respectively. The computing unit is configured to automatically classify power outlets within the perception data to assign the power outlets to one of the plurality of power cables that is compatible with the power outlet within the perception data.

[0111] In another embodiment, the computing unit is configured to: a.) further classify the compatible power outlet based on its charging rate characteristics, in particular based on voltage and current; and b.) dispatch the robot to the location of the power outlet, in particular to the reachable power outlet with the highest maximum charging rate, further based on the charging rate characteristics of the power outlet.

[0112] In another embodiment, the system comprises a plurality of mobile robots. Each robot of the plurality of mobile robots comprises a power cable and a power outlet. The power cable and the power outlet of the robot are compatible with each other such that two of the robots can be connected between each other and electric energy can be transferred between the two robots. One robot of the plurality of mobile robots is configured to send an alarm signal if it cannot access a charging station nor a power outlet. The system is configured to dispatch another robot of the plurality of mobile robots to the location of the one robot. The one robot and the another robot are configured to connect to each other autonomously by connecting the power cable of the one robot to the power outlet of the another robot to transfer electric energy from the another robot to the one robot. BRIEF DESCRIPTION OF DRAWINGS

[0113] In the following, different aspects of mobile robots and systems according to the present application are described or explained in more detail by way of example only, with reference to the working examples schematically illustrated in the accompanying drawings. Identical elements are marked with the same reference signs in the drawings. The embodiments are generally not shown to scale and these embodiments should not be interpreted as limiting the present application. In particular:

[0114] Figure 1 : Exemplary embodiment of a humanoid robot according to the present application;

[0115] Figure 2 : Another exemplary embodiment of a humanoid robot according to the present application, wherein the robot comprises track drives for a standing mode and a full-track mode;

[0116] Figure 3 : Several exemplary embodiments of a wheel arrangement as can be implemented in a robot according to the present application;

[0117] Figure 4 : Exemplary embodiment of a mechanically actuated leg as can be used in a robot according to the present application, wherein the leg comprises two drive shafts and a transmission element such that one of the drive shafts provides mechanical actuation of the other drive shaft via the mechanically actuated element;

[0118] Figure 5: Exemplary application of a robot for monitoring an environment according to the present invention;

[0119] Figure 6 : Another exemplary embodiment of an anthropomorphic robot according to the present invention, wherein the robot is configured to perform coordinate measurement of a measurement object by tactile probing;

[0120] Figure 7 : Another exemplary embodiment of an anthropomorphic robot according to the present invention, wherein the robot is configured to perform coordinate measurement of a measurement object by optical probing;

[0121] Figure 8 : Another exemplary embodiment of an anthropomorphic robot according to the present invention, wherein the robot comprises a coupling interface configured to provide a releasable rigid docking to a measurement platform;

[0122] Figure 9 : Exemplary application of a robot according to the present invention in a factory environment with multiple machines, wherein the robot is configured to use sound data to pick out noise from the machines to monitor a wear state of the machines;

[0123] Figure 10 : Figure 9 : A robot of the present invention, wherein the robot has approached a machine to perform a vibration measurement for determining a wear state of the machine;

[0124] Figure 11 : Another exemplary embodiment of an anthropomorphic robot according to the present invention, wherein the robot comprises a power cable compatible with a power outlet of an environment and is configured to autonomously find the power outlet and attach the power cable to the power outlet;

[0125] Figure 12 : Another exemplary embodiment of an anthropomorphic robot according to the present invention, wherein the robot comprises a power cable and is configured to autonomously connect the power cable to another robot;

[0126] Figure 13 : Exemplary embodiment of a robot drive module as can be used in a robot according to the present invention;

[0127] Figure 14 : Exemplary embodiment of a dual shaft drive motor is depicted as can be used in a robot according to the present invention. DETAILED DESCRIPTION

[0128] Figure 1An exemplary embodiment of an anthropomorphic robot 1 according to the present application is shown. The robot 1 comprises a main body 2, two mechanically actuated leg portions 3, two mechanically actuated arm portions 4, and a head portion 5, for example wherein the head portion is also mechanically actuated for being movable relative to the main body 2.

[0129] The two leg portions 3 are attached to the main body 2 at a lower portion of the main body, and are configured to provide locomotion of the robot 1 over a ground surface. Each of the leg portions 3 is attached to the main body 2 by a hip joint 6, which provides movement of an upper portion 7 of the leg portion relative to the main body 2. Each leg portion 3 further comprises a knee joint 8, which provides movement of a lower portion 9 of the leg portion relative to the upper portion 7 of the leg portion, and a wheel 10 at a distal end remote from the knee joint 8 for contacting the ground surface to provide locomotion. For example, each of the hip joints 6 provides two degrees of rotational freedom of movement relative to the main body 2, and each of the knee joints 8 provides one degree of rotational freedom of movement, for example to provide a folding movement of the leg portion 3.

[0130] The two arm portions 4 are attached to the main body 2 at an upper portion of the main body 2 and are configured to move relative to the main body 2. Each of the two arm portions 4 is attached to the main body 2 by a shoulder joint 11, for example wherein the shoulder joint 11 provides two degrees of rotational freedom of movement relative to the main body. In the shown example, each of the two arm portions 4 further comprises an elbow joint 12 and a hand joint arrangement 13. Each of the elbow joints 12 provides one degree of rotational freedom of movement, for example for providing a folding movement of the arm portion. The hand joint arrangement 13 is configured for performing grasping operations. As an example, the hand joint arrangement is configured to provide a hand that is embodied like a human hand. Alternatively, the hand joint arrangement provides a more simple grasping tool, like two or three finger paws.

[0131] The head portion 5 comprises a top side 14 opposite an attachment point 15 of the head portion 5 to the main body 2, a front side 16 adjacent to the top side 14, and a rear side 17 opposite the front side 16 and adjacent to the top side 14, and two opposite lateral sides 18, 19 adjacent to the top side 14 and to the front side 16 and the rear side 17, respectively. In other words, assuming a nominal head position of the robot during forward movement on a horizontal ground surface, the top side 14 faces the sky, the front side 16 (also referred to as the “face” of the robot) faces the direction of travel, the rear side 17 faces backwards, and the two lateral sides 18, 19 face in a horizontal direction perpendicular to the direction of travel.

[0132] The robot 1 is configured to provide locomotion in a walking mode by stepping movements of the leg portions 3, and in a driving mode by rolling on the wheels 10. The driving mode is also referred to as a so-called “Segway” mode, wherein the robot 1 is configured to provide self-balancing movement control of the wheels 10 to maintain the robot in a defined upright position, for example by means of a control algorithm that makes use of inertial sensors and gyroscopic sensors to automatically provide motor adjustments to maintain the upright position.

[0133] For example, the robot is further configured to provide self-balancing movement by means of automatic movement control of the movement of the arm portions during the travelling such that the movement of the arm portions 4 supports the robot to balance in a defined upright position. As an example, the robot comprises a control algorithm configured to automatically adjust the relative attitude of the two mechanically actuated arm portions based on the tilt sensor and the gyro sensor such that the weight and orientation of the arm portions 4 provide a stabilizing effect to maintain the upright position of the robot 1.

[0134] The robot can be configured to select the appropriate travel mode by itself, for example based on an analysis of the ground and / or the environment. By way of example, the step motion provides movement over uneven terrain or movement walking up stairs. The travel mode can further be configured to provide a curve movement, wherein the robot is “tilted” into a curve by shortening one of the leg portions, for example to allow to take curves at a faster speed and / or to reduce the curve radius.

[0135] The robot 1 can be configured to provide a storage configuration by folding the arm portions 4 and the leg portions 3, wherein in the storage configuration the sum of the height, width and length and / or the volume of the enveloping cuboid of the humanoid robot 1 is reduced. For example, the robot 1 is configured to assume the storage configuration without assistance by using a defined sequence of control of the motors driving the joints of the robot.

[0136] Although some of the aspects described herein have been explained with reference to a more simple type of mobile robot, it goes without saying that these aspects can also be implemented based on a humanoid robot 1 as described above.

[0137] In the embodiment shown in the figures, the robot 1 further comprises various different sensor types and a specific embodiment of a power supply using a replaceable battery, as described below with reference to Figure 1 The different embodiments of the humanoid robot according to the present application can comprise different combinations of these sensor types (for example, wherein only one of these sensor types is used) and different power supplies, respectively.

[0138] Here, one of the two leg portions 3 comprises a battery compartment 20 arranged between the knee joint 8 and the hip joint 6. The battery compartment 20 is configured to accommodate a replaceable battery such that when the battery is accommodated in the battery compartment 20, the battery provides electrical energy for driving the movement of the two leg portions 3 and the two arm portions 4. The robot 1 is configured to provide a battery replacement for the battery compartment 20 during the continuous operation of the robot, for example during the movement of the two arm portions 4.

[0139] In particular, the robot 1 comprises in each leg such a battery compartment 20 that both are arranged between the respective knee joint and the hip joint. Each of the battery compartments is configured to house a battery that provides electrical energy for driving the movement of the two legs and the two arms. Each of the battery compartments comprises a circuit breaker that is configured to be activated for battery replacement of the respective battery compartment and provides electrical disconnection of the battery located in the respective battery compartment so that the robot is still supplied with electrical energy from the other battery compartment. For example, the robot 1 is configured to autonomously replace the battery of one of the battery compartments, e.g. by using one of the arms 4. The quick release function provided by the battery compartments 20 can help the robot to perform autonomous replacement of the battery, e.g. where the quick release is activated by manipulation of the arms or hands of the robot and / or where the quick release is activated by an electronic signal provided by the robot 1.

[0140] One of the arms 4 comprises an optical sensor 21 that is configured to provide optical sensor data, e.g. visual data provided by a camera or laser-based data provided by a laser radar unit. The hand joint device 13 is configured to provide a human-like hand with five fingers and the optical sensor 21 is arranged at the tip of one of the five fingers. For example, the robot 1 is configured to use the optical data for optical probing of a surface to be measured. Alternatively or in addition, the robot 1 is configured to use the optical sensor data to assist a gripping operation of the hand.

[0141] The robot 1 further comprises a tactile sensor 22 arranged at one of the other fingers. For example, the tactile sensor 22 comprises a ruby ball and the mobile robot is configured to provide tactile scan data by providing coordinates of the distal end point of the tactile sensor 22.

[0142] The head 5 comprises a scanning laser radar unit 23 that is configured to provide scanning movement of a laser measurement beam relative to two rotational axes 24, 25 during movement of the robot on the ground. For example, one of the two rotational axes 25 of the laser radar unit 23 passes through the front side 16 and the rear side 17 and the laser radar unit 23 is configured to provide a front view field of view around one of the two rotational axes 25 and, based thereon, to generate light detection and ranging data for generating a three-dimensional point cloud.

[0143] Here, the head 5 further comprises a ToF unit 26 that comprises an arrangement of time-of-flight sensors and is configured to provide 3D imaging data of the environment during movement of the robot on the ground to generate a 3D model of the environment. In the shown embodiment, the ToF unit 26 is arranged at the front side 16 and provides a front view field of view.

[0144] The head 5 further comprises two imaging cameras 27, namely one camera arranged on each of the lateral sides 18, 19. The cameras 27 are arranged opposite to each other and provide opposite "side view" fields of view. For example, each camera 27 is a color camera, e.g. an RGB camera, which is configured to provide a complete spherical 360° field of view. The cameras 27 can also be embodied as so-called RIM (range imaging) or ToF (time of flight) cameras.

[0145] The body 2 comprises a structured light scanner 28, which is for example arranged in a chest region in an upper region of the body 2 and which is configured to provide a front view field of view. As an example, the structured light scanner 28 is used for 3D scanning of the environment during movement of the robot 1 on the ground and / or for 3D scanning of specific objects in the environment, e.g. for which the robot is placed in front of a specific object to be measured with the chest facing the object. For example, for such object scanning, the robot 1 can take one or more positions around the object with always the chest facing the object to provide a full-compass scan of the object.

[0146] The robot 1 can comprise further sensor types, e.g. a positioning unit 29 (e.g. to provide position and / or navigation), a communication unit (not shown), a microphone unit 30, a communication unit (not shown), an inertial measurement unit (not shown) and a thermal imaging unit 31 unit. As an example, the positioning unit 29 is embodied as a GNSS-based or WLAN-based positioning unit arranged in the head 5, the microphone unit 30 is also arranged in the head 5 and the thermal imaging unit 31 is arranged in the chest region of the body 2.

[0147] As an example, the robot 1 further comprises a simultaneous localization and mapping (SLAM) unit which is configured to perform a simultaneous localization and mapping process based on perception data provided by one of the above-mentioned sensors. For example, the SLAM process is based on visual data provided by the cameras 27 in the head 5 and / or on lidar data provided by the lidar unit 23 in the head 5.

[0148] As an example, a robot according to the present application, e.g. a humanoid robot 1 as described above, can be configured for hidden feature inspection, as it can move substantially freely at any time and to any location where an inspection is required. Using local inspection sensors (e.g. probe sensors and camera / laser based tracking sensors, like Figures 6 to 8The robot can reach geometrical features to be inspected that are conventionally not reachable without repositioning the measuring machine relative to the object to be measured, which is often quite cumbersome. The robot can easily move to the backside of a part to be inspected, or even to the inside of a part to be inspected (at least partially), and thus provides greater flexibility and autonomy for inspecting parts. In particular, the robot is particularly suitable for inspecting large parts.

[0149] For example, measurements by the local sensors in the hand (see for example Figures 6 to 8 ) in combination with an accurate robot-based position of the hand / local sensors. By determining the position of the robot within the inertial system, for example by using a fiducial tracker or by taking a measurement with one hand relative to the robot and referencing the other hand, large parts can be inspected flexibly, such as wind turbines, blades, power plant parts, ship parts, airplanes, trains, etc. As an example, large parts are inspected during assembly, where some local features need to be inspected accurately and with a certain accuracy related to the external inertial system, for example fitting a wing on an airplane. The robot can move to the position where the interfaces of the parts to be fitted are located, inspect them with the local sensors, and obtain the accurate relationship between the features on both sides of the interface by placing one hand on one feature (for reference) and the other hand on the other feature to be inspected and / or by tracking the global position of the robot and simultaneously using the local sensors on the hands for the inspection.

[0150] Figure 2 Another exemplary humanoid robot 101 according to the present application is depicted, here with an alternative embodiment providing locomotion of the robot 101. The legs 3 are attached to the body 2 by hip joints 6. Each of the legs 3 comprises an upper part 7 (in particular an upper part 7 comprising a battery compartment (not shown, see Figure 1 ) and a lower part 9 connected to each other by a knee joint 8, wherein the knee joint 8 provides a pivotal movement of the lower part 9 relative to the upper part 7. The lower part 9 comprises a track 32 running over a lower pulley 33 and an upper pulley 34. The lower pulley 33 is arranged distally away from the knee joint 8. The upper pulley 34 is arranged closer to the knee joint 8, for example wherein its suspension arrangement can comprise elements of the knee joint 8. By way of example, as shown, the complete lower part 9 is embodied as a track 32.

[0151] The lower part 9 of the leg (track 32) further comprises one or more support pulleys 35 arranged between a lower pulley 33 and an upper pulley 34 such that the one or more support pulleys 35 are in contact with the track 32. For example, the upper pulley 34 is a driven pulley, wherein the upper pulley 34 is driven by making use of components arranged in the knee joint 8 and / or the upper part 7 of the leg. The lower pulley 33 can also be a driven pulley, for example wherein separate motors drive the upper pulley 34 and the lower pulley 33, respectively. For example, the driven pulleys comprise electric motors integrated into or on the pulleys.

[0152] The left side of the figure depicts the robot 101 in a so-called "standing mode", while on the right side the robot 101 is depicted as being in a so-called "full track mode".

[0153] In standing mode, the surface contact area of the running surface has a smaller area size than the surface contact area of the running surface provided by the full track mode. In the example shown, the robot 101 supports itself by standing on a curved part of the track 36, which is curved through a circumferential area of the lower pulley 33, wherein the travel is provided by self-balancing travel (on the lower pulley 33). As an example, the robot comprises an inertial sensor unit 38 in the main body 2, for example a tilt sensor and a gyro sensor, and a control algorithm configured to automatically control the movement of the lower pulley 33 based on the sensor readings such that the robot balances in a defined upright position.

[0154] For example, in full track mode, the robot 101 supports itself by standing on the largest possible surface contact area provided by the track 37, and the travel is provided by track travel. The track drive can comprise a further inertial sensor 39, for example an accelerometer. For example, the further inertial sensor 39 in the track drive provides wheel slip, track slip or impact information.

[0155] In another embodiment, the robot is configured to provide a lifting of the upper pulley relative to the lower pulley during travel to provide a selection between two different running surface contact areas, wherein the two contact areas differ from each other by their respective area size. In other words, the robot is configured to provide different "foot sizes" by lifting the upper pulley to different heights above the ground and being able to maintain different lifted positions during travel. For example, the robot is configured to actively or passively change the running surface contact area (change the area size), for example by comprising and using additional "displaceable pulleys" (at least one per track drive) and spring preloading to obtain preloaded tracks. As an example, the robot can adjust the "foot size" depending on the task of the robot, for example adjusting the lifted position and using a small foot size when standing in place and keeping the position is important, but using a more flat foot size (larger foot size) when climbing steep and difficult terrain.

[0156] Passive variants can also be designed with springs / dampers, which can then be designed to be semi-active (change spring constant or damping properties depending on the task of the robot).

[0157] In one embodiment, the robot 101 uses optical sensor means, such as stereo cameras and / or laser radar units arranged in the head 23, to provide a view of the ground in the direction of motion of the robot. The optical sensor means provide a prediction of ground state information about the subsequent path to be traveled by the robot 101, for example, wherein the ground state information provides the hardness and roughness of the ground and / or wherein the ground state information provides geometric information such as slope information, and / or wherein the ground state information provides information about obstacles in the path. Additional sensors, such as inertial sensors in the track drives 39, can provide data that contribute to the ground state information. For example, the robot 101 is configured to automatically change between full-track mode and stand mode based on the prediction of the ground state information.

[0158] In addition, the robot 101 can provide locomotion by stepping motion of the leg 3, for example, wherein the lower pulley 33 is locked during the stepping motion of the leg. Alternatively, the lower pulley 33 can be idling.

[0159] It goes without saying that the robot can comprise various types of damping elements that reduce the excitation of the resonance frequency or at least make at least one of the stepping locomotion, the rolling locomotion and the track locomotion more robust. The damping elements can be added in parallel or in series to any structure and relative moving element. They can be passive, semi-active or active.

[0160] Figure 3 Several exemplary embodiments of wheel arrangements that can be implemented in a robot according to the invention are depicted, namely a single wheel 10 arranged at the distal end of the lower part of the leg 9 (on the left), a two-wheel arrangement with two wheels 10 mounted to a suspension element 41 (in the middle), and a three-wheel arrangement 43 (on the right).

[0161] As an example, the suspension elements are mounted to the main shaft of the respective wheel arrangement, such that they are at least rotatable relative to the main shaft 42. The suspension elements 41 can be rigidly connected to each other or can be independently rotatable. For example, the suspension elements 41 or the main shaft 42 comprise an elastic material.

[0162] For example, the arrangement with multiple wheels provides improved traversal of uneven ground by providing two stable contact points for each leg 3.

[0163] The three-wheel arrangement 43 comprises three wheels 10 attached to a wheel carrier 44 configured to support the three wheels 10 in a circular pattern, e.g. where the wheel carrier 44 is rigid. A main shaft 42 connects the wheel carrier 44 to the rest of the leg 3 and provides a rotational degree of freedom for the wheel carrier 44. Three wheel shafts 45 connect the wheels 10 to the wheel carrier 44 such that each wheel shaft 45 provides a wheel rotational degree of freedom independent of the rotational degree of freedom of the main shaft and the wheel rotational degrees of freedom of the other two wheel shafts. As an example, in addition to improved stability, the three-wheel arrangement provides for stair climbing via rolling locomotion, which can be advantageous over step-wise motion, e.g. when carrying heavy loads.

[0164] Figure 4 An exemplary embodiment of a mechanically actuated leg as can be used in a robot according to the present application is depicted with dashed lines. The leg comprises an upper part 7 and a lower part 9 such that the upper part 7 and the lower part 9 are connected by a knee joint 8 configured to provide a pivotal movement of the lower part 9 relative to the upper part. The upper part 7 of the leg comprises a drive shaft 46 mechanically connected to an electric motor 47. The knee joint comprises a further shaft 48 mechanically connected to the knee joint such that a rotational and / or pivotal movement of the further shaft 48 causes a pivotal movement of the knee joint. The drive shaft 46 and the further shaft 48 are connected by a mechanical transmission element 49 such that the drive shaft 46 provides a mechanical actuation of the further shaft 48 via the mechanical transmission element 49. In the depicted embodiment, each of the drive shaft 46 and the further shaft 48 comprises a toothed region and the mechanical transmission element 49 is a chain transmission. Alternative embodiments can comprise (a non-exhaustive list) a belt drive, e.g. a V-belt, or a rack as the mechanical transmission element 49.

[0165] As an example, the electric motor 47 is arranged at the hip and is configured as a so-called dual shaft drive motor comprising two nested motors, where each of the two motors can be rotated at a speed and in a direction independent of the other. This provides a compact drive module configured to independently drive the hip joint 6 and the knee joint 8. For example, the dual shaft drive motor is configured as described with reference to Figure 14 .

[0166] Figure 5 An exemplary application involving a robot according to the present application for monitoring an environment, e.g. a commercial complex. The environment comprises movable barrier objects, such as a door 50 and a window 51. In the shown example, the door 50 is closed and the window 51 is in a partially open position. Such a condition can lead to a safety risk, e.g. a burglary, potential equipment damage due to adverse weather conditions or additional heating needs.

[0167] The depicted environment also comprises a general security camera 54 mounted on the ceiling. While such general security cameras 54 are suitable to provide information about the general condition of the area, they are less suitable to provide fine details about the condition of the obstructing objects 50, 51, especially due to their fixed position, e.g. to assess whether the obstructing objects are properly closed and / or secured. To overcome this deficiency, a mobile robot, for simplicity only the head 5 and the hand joint device 13’ of the robot are depicted in the figures, is used to patrol the environment to control the state of the environment, in particular the state of the movable obstructing objects 50, 51.

[0168] The exemplarily depicted robot comprises a wireless communication module 55 for accessing a digital reference map of the environment comprising a set of movable obstructing objects. The patrol path of the robot is derived to provide a passage by each of the known movable obstructing objects. It is also foreseen that the robot locates itself in the environment and identifies previously unknown movable obstructing objects.

[0169] The robot further comprises a laser radar unit 23 in the head, wherein the laser radar unit 23 is configured to provide a coordinate scan of the environment, and the robot is configured to use the laser radar data of the laser radar unit 23 to automatically identify the movable obstructing objects 50, 51 within the environment. In the schematically provided graph by line scans, the laser radar data comprises information about the blocking state of the movable obstructing objects. For example, the laser radar scan is specifically used to obtain information about the blocking state of the movable obstructing objects 50, 51. Alternatively, the laser radar scan to obtain information about the blocking state is a byproduct of a continuously performed survey scan to obtain 3D information about the surrounding environment, e.g. to provide navigation information for the robot.

[0170] The blocking state of the movable obstructing objects 50, 51 is determined via geometric test parameters 58, depicted in the figures as distance jumps, and corresponding test criteria. As an example, the geometric test parameters comprise angle information or distance information. Further, the robot is configured to identify control components, such as window handles 52 and door handles 53, wherein the geometric test parameters 58 and associated test criteria are used to correlate the geometric pose of the control components 52, 53 with the locking state of the movable obstructing objects 50, 51. As an example, further control components that the robot can identify are sliding or rotating latches, mechanical door locks, padlocks, etc. Further, some control components can comprise a visible blocking element, e.g. a sliding or rotating latch, and the geometric test parameters provide the geometric pose of said blocking component. For example, the robot is configured to access a database comprising control components 52, 53 associated with a given movable obstructing object 50, 51 and the nominal position of the control components 52, 53 relative to the movable obstructing object associated with an open or locked state.

[0171] As an example, the robot is further configured to access a 3D reference model representing the environment with the obstacle objects 50, 51 in a reference state, wherein the robot is configured to determine the alarm state based on a difference analysis for identifying differences between the 3D reference model and a 3D model of the environment determined from actually generated lidar data. For example, the difference analysis is performed to compare the geometric pose of the movable obstacle objects 50, 51 or control components 52, 53 with the pose in the reference model. Then, the robot transmits the alarm state to the on-site supervision unit via the wireless communication module 55.

[0172] In particular, the robot can be configured to provide time-stamped lidar data and to determine the target state of the blocking state of the obstacle objects 50, 51 based on the time stamp. For example, the target state corresponds to a defined schedule providing different target states depending on the time of day, e.g., wherein different target states are determined for day and night. The robot can be further configured to provide a learning of different target states based on an evaluation of a 3D model of the environment generated based on older lidar data.

[0173] The robot can be configured to autonomously patrol the environment according to a predefined patrol path. This can include a specific deployment of the robot to check for target obstacle objects. For example, the robot is configured to provide a user interface allowing to select one or more obstacle objects 50, 51. Upon receiving an investigation request, e.g., including 1.) navigating to a target location provided by the investigation request; 2.) identifying movable obstacle objects in the surrounding of the target location; 3.) determining the blocking state of the movable obstacle objects 50, 51 in the surrounding of the target location; 4.) and reporting the blocking state of the movable obstacle objects 50, 51 by the wireless communication module 55, the robot autonomously performs the tasks of the request.

[0174] For interacting with the environment, e.g., opening / closing a door 50 or a window 51, the robot comprises a hand joint device 13’ with a mechanically actuated gripping element 60. In the shown embodiment, the gripping element 60 forms a three-fingered claw. Various other configurations are possible, e.g., depending on the tasks of the robot and the type of objects to be interacted with. By way of example, the gripping element 60 further comprises a specific interaction element configured to provide an interaction with a specific object. For example, the specific interaction element is configured to provide an interaction with a key or keyhole, a frame or ball locking element, a magnetic pairing device, or a tool. The robot is configured to intervene in the control component 52, 53 of the obstacle object 50, 51, in particular to intervene the gripping element 60 to change its blocking state.

[0175] The robot is configured to autonomously interact with the barrier object 50, 51, e.g. upon detecting a deviation of the barrier state of the barrier object 50, 51 from a target state. In particular, the robot can be configured to autonomously intervene with the barrier object 50, 51 based on a warning, e.g. wherein the warning is based on data of the robot or external sensors. For example, the warning is based on temperature or humidity data of the environment, and the robot is configured to close an open window 51 in case the temperature is below a threshold / the humidity is above a threshold. The warning can also be based on weather forecast data including a warning for adverse weather conditions and the robot, e.g. wherein the robot is configured to verify the resulting barrier state of the barrier object 50, 51 and to open / close the barrier object based on given weather-related rules.

[0176] Figure 6 and Figure 7 An embodiment of a robot according to the present application is depicted, wherein the robot is configured to perform a coordinate measurement of a measurement object 61 by tactile probing measurements ( Figure 6 ) or optical probing measurements ( Figure 7 ) performed with sensors arranged in the hand of the robot.

[0177] For example, as Figure 6 depicted, the robot comprises a hand joint arrangement 13 forming a hand with five fingers, wherein one of the fingers is configured to act as a tactile sensor unit 22. The coordinate measurement is performed by bringing the tactile sensor unit 22 into contact with the measurement object 61 and by guiding the tip of the tactile sensor along a measurement path 64 on the object 61 while adjusting the pose of the joints in recording the coordinates of the measurement path 64.

[0178] For example, the depicted shoulder joint 11 provides two rotational degrees of freedom of the shoulder joint, the depicted elbow joint 12 provides one rotational degree of freedom, and the hand joint arrangement 13 provides two further rotational degrees of freedom.

[0179] The coordinates are determined by using angle encoder data provided by a plurality of angle encoders 68, 69, 70 configured to measure the angular position of the respective joint 11, 12, 13. While more degrees of freedom result in more complex measurement paths 64, inaccuracies of the angle encoders 68, 69, 70 and / or driving components of the joints 11, 12, 13 result in a reduced precision of setting the measurement path 64. For example, the angle encoders 68, 69, 70 are provided by a robot drive module, as shown in Figure 13 , e.g. to provide the necessary angular position accuracy.

[0180] In the illustrated embodiment, the arm portion 4 further comprises an optical sensor 21 arranged or integrated into the hand of the robot. The optical sensor 21 can be a metrology grade sensor for providing probe measurements by itself (see e.g. Figure 7 ). Here, however, the optical sensor unit 21 is configured as a vision aid to ensure that the tactile sensor unit 22 is in proper contact with the measurement object 61 or that the measurement path 64 does not contain obstacles. For example, such a vision aid is advantageous in case the field of view of the measurement path 64 is obstructed (as seen from the optical sensor on the head or chest of the robot).

[0181] Figure 7 Another embodiment of a mobile robot 201 according to the present application is depicted, wherein the robot 201 is configured to perform coordinate measurements on a measurement object 61 by using an optical sensor unit 21 integrated into the hand of the robot or arranged in the hand of the robot. The optical sensor unit 21 is configured to perform laser-based distance probing, e.g. based on laser interferometry, time-of-flight timing (measuring the start and stop times of emitted and returned probe signals) or laser triangulation. The optical sensor unit 21 can also be configured to perform camera-based distance probing, e.g. based on photogrammetric methods such as stereo imaging.

[0182] In the illustrated embodiment, the robot 201 comprises a tracking device 71 arranged in the head and a tracking device 72 arranged in the main body 2, e.g. wherein the tracking devices are embodied as laser trackers. The tracking devices 71, 72 are used to determine and track a reference position close to the optical sensor, which for example allows to reduce the requirements on the positioning accuracy of the angle encoders associated with the joints up to the reference position. Alternatively or additionally, the tracking sensors 71, 72 are used to calibrate the angle encoders of the arm portion.

[0183] In the illustrated embodiment, the robot 201 further comprises a camera 27 arranged in the head, e.g., which is configured to provide image data for positioning a certain region of the arm portion, e.g. for image-based tracking of the arm portion movement and (indirectly via the arm portion movement) of the optical sensor unit 21 based on image. For example, to assist the image-based tracking, the arm portion comprises a tracking pattern 74 and the robot 201 is configured to automatically identify the tracking pattern portions 74 in the image by the camera 27 and to associate one of the pattern portions 74 with the relative position of one of the pattern portions with respect to the optical sensor unit 21. In particular, a plurality of tracking patterns 74 is provided to enable tracking of the pose of a plurality of joints and / or arm segment components. For example, the use of the tracking pattern 74 allows for uninterrupted tracking in case the tracking sensors 27, 71, 72 do not have a direct line of sight on the tactile sensor unit or the optical sensor unit.

[0184] Instead of or in addition to using tracking patterns or tracking targets, the robot is further configured to access a 3D model of the arm portion and to identify arm portion features within the image of the camera 27. The robot can then automatically associate one of the arm portion features with an arm portion feature within the 3D model of the arm portion and, based thereon, derive the relative position of the one of the arm portion features with respect to the optical sensor unit 21 or the haptic sensor unit 22.

[0185] As an example, the robot 201 is itself further configured to be tracked, e.g. for referencing the robot’s measurements in the external coordinate system, for which the robot 201 comprises a tracking target 73 on each of the left and right hand joint arrangements 13. The tracking target 73 is embodied as a retroreflector, for example, and is foreseen to be measured and tracked by an external laser tracker, wherein the tracking target is configured such that the external laser tracker can determine the pose of the tracking target with six degrees of freedom (6DoF). The optical sensor unit 21 has a fixed, rigid relative pose with respect to the tracking target 73.

[0186] Figure 8 A further embodiment of a robot according to the present application is depicted, wherein the robot 301 is configured to perform a coordinate measurement of a measurement object 61 by means of a haptic sensor unit 22 arranged on the hand of the robot 301. Here, the measurement object 61 is placed on a measurement platform 75 comprising a first docking point 76 and a second docking point 77. The robot 301 comprises a coupling interface 78 configured to provide a releasable rigid docking to the docking points 76, 77 of the measurement platform 75. Due to the rigid docking, any movement of the haptic sensor unit 22 with respect to the robot 301 only depends on the movement of the joints between the haptic sensor unit 22 and the coupling interface 78. Thus, it is sufficient to provide an accurate angular positioning determination only for this subset of joints between the haptic sensor unit 22 and the coupling interface 78.

[0187] As an example, each joint of this subset of joints is equipped with a robot drive module as Figure 13 depicted.

[0188] Alternatively or in addition, the robot 301 comprises a laser tracker 72 in its body. The laser tracker 72 is rigidly mounted on the robot in a fixed relationship to the position of the coupling interface 78. The laser tracker is configured to provide tracking of a part of the robot arranged between the coupling interface 78 and the haptic sensor unit 22. For example, the laser tracker is used to track the haptic sensor unit 22 and / or a part of the arm portion 4 comprising the haptic sensor unit 22. The tracking by the laser tracker 72 can then be used to determine 3D position data of the tracked part of the robot, e.g. wherein the tracked part of the robot comprises a tracking target for assisting laser-based tracking.

[0189] In the illustrated implementation, the robot 301 further comprises a camera 27 located in the head portion. While the head portion does not have a fixed positional relationship to the coupling interface 78, it can comprise an end stop having a fixed positional relationship to the coupling interface 78. Thus, similar to the laser tracker 72, the camera 27 can be used to provide automated image-based tracking of a part of the robot arranged between the coupling interface 78 and the haptic sensor unit 22. For example, the tracking by the camera 27 is assisted in that the part to be tracked comprises a visual pattern for assisting image-based tracking.

[0190] Another arm portion of the robot comprises another hand joint arrangement. The other hand joint arrangement is configured to grasp an object, e.g. the measurement object 61. Similar to the arm portion comprising the haptic sensor unit 22, any movement of the hand of the other arm portion only depends on the movement of the joint between the haptic sensor unit 22 and the coupling interface 78.

[0191] As an example, the docking points 76, 77 are associated with a docking point pattern 79, e.g. embodied as a barcode. The robot 301 is configured to use the images of the camera 27 for pattern recognition and pattern analysis of the docking point pattern 79 associated with one of the docking points and to associate the docking point pattern 79 with pose information of the respective docking point. For example, the robot is configured to access a database comprising different patterns and associated pose parameters for each of the different docking point patterns 79. Alternatively or in addition, the docking points 76, 77 and the coupling interface 78 can be configured for transmitting the pose information by electronic communication through the docking points 76, 77.

[0192] For the depicted measurement platform 75 with two docking points 76, 77, the robot can be configured to perform the distance probing measurement such that: 1.) the robot 301 docks to the first docking point 76, 2.) a first distance probing measurement is performed by the tactile sensor unit 22 relative to the first docking position 76, 3.) the robot 301 docks to the second docking point 77, 4.) a second distance probing measurement is performed by the tactile sensor unit 22 relative to the second docking position 77, and 5.) the first distance probing data is merged with the second distance probing data based on pose information of the first docking point 76 and the second docking point 77.

[0193] The pose information can be obtained via electronic communication. The relative pose of the first docking point and the second docking point can be obtained via merging the essential features 80, 81 of the measured object 61 in the first distance probing data and the second distance probing data. Essential features 80, 81 in the sense of the present invention are accessible from the first docking point 76 and the second docking point 77 and are identifiable in the distance probing data due to visual, geometric or other distinguishing characteristics.

[0194] As an alternative (or in addition) to the docking mechanism for rigid docking, the coupling interface comprises a high-precision coupling sensor unit, for example comprising capacitive sensors with a measurement range of up to several millimeters, which are distributed in space and measure the position 6DoF changes relative to the coupling counterpart once the coupling interface is arranged in the docking area. As an example, visual video cameras or other sensors can be used to measure influences in the vicinity of the docking area.

[0195] The docking mechanism (i.e. the coupling interface) can be placed anywhere on the robot, for example on the hand, at the body or at the arm segment. For example, a plurality of coupling interfaces are provided on the robot.

[0196] Figure 9 An exemplary application of a robot according to the present invention in a factory environment with a plurality of machines 82, 83 is depicted. The robot comprises a microphone unit with a plurality of microphones 30, in particular at least three microphones, configured to acquire sound produced within the factory environment. A first machine 82 produces a noise 84 deviating from the norm. The robot comprises an event detector / localization unit 85 to identify the noise 84 deviating from the norm and to determine a location 86 of the isolated noise 84. The event detector / localization unit 85 can utilize the phase shift between the three microphones 30 to localize the source of the isolated noise. For example, the robot comprises unidirectional microphones, for example low-sensitivity microphones with a sensitivity of less than 40 dBV, and the event detector / localization unit 85 determines the location 86 of the isolated noise 84 based on the orientation and sound level of the unidirectional microphones.

[0197] The robot is configured to derive a path 87 to the location 86 of the isolated noise 84 based on a digital model of the environment derived by the robot's perception sensors. The digital model can include the locations of other machines. In particular, the robot is configured to avoid the further machines and other obstacles 89. The event detector / localization unit 85 can have access to further data, in particular from a plurality of microphones installed in the factory environment in fixed positions and / or associated with specific machines 82, 83. Alternatively, the path 87 is determined by using an external device and provided to the robot, and the robot is configured to receive the path 87 via the wireless communication module 55.

[0198] As an example, the robot is configured to derive a patrol mode based on the digital model of the environment, such that at least one sound collection point of the patrol mode, sound produced by each of the plurality of machines 82, 83 is collected. Alternatively or in addition, such a patrol mode is provided by means of an external device, and the robot is configured to receive the patrol mode via the wireless communication module 55.

[0199] As an example, the system is configured to train a vibration model using a machine learning algorithm, which provides an indicator for different maintenance states of the machine, e.g. by training the vibration model with normal / abnormal data and then using it with measured vibration signals to identify whether an abnormal state is present. For example, the training of the vibration is combined with operator feedback, and the vibration model is retrained based on the operator feedback.

[0200] Similarly, the detection of the isolated noise can also be trained based on normal / abnormal states of the factory, and continuously retrained to improve the quality of the isolated noise event detection.

[0201] Figure 10 Reference is made to Figure 9 The described robot 401, wherein the robot 401 has approached a first machine 82 producing noise 84 deviating from the norm. The robot 401 comprises a vibration sensor 90 arranged on a mechanically actuated arm 4. By autonomously bringing a probing part of the vibration sensor 90 into contact with a vibration measurement point 91 on the machine 82, the robot 401 derives vibration data of the machine 82. For example, the vibration sensor 90 comprises at least one of an inertial measurement unit, a microphone, a piezoelectric sensor, and an optical sensor. In particular, the vibration sensor 90 is configured to provide the vibration data by direct motion coupling.

[0202] As an example, the vibration measurement points 91 are predefined points on the machine, e.g. where the robot is configured to access a lookup table to autonomously identify the vibration measurement points 91 on the machine 82. The measurement points 91 can be specially marked on the machine to help the robot 401 to find them, or the robot can be configured to find the measurement points 91 by analyzing the geometry of the machine. The database can further comprise reference data for normal operation of the respective machine 82 and / or for specific wear / fault states of the respective machine 82. For example, the database comprises vibration spectra representing said different states.

[0203] Figure 11 Another embodiment of a robot 501 according to the invention is depicted, wherein the robot 501 comprises a power cable 93 that is compatible with a power outlet 92 of the environment, and one arm portion is configured to autonomously plug the power cable 93 into the power outlet 92. The robot 501 is configured to derive a model 88 of the environment based on perception data, e.g. provided by a camera, a lidar unit and / or a structured light scanner, and to localize 94 itself in the model 88, e.g. based on a SLAM process. The robot 501 is further configured to identify the power outlet 92 in the environment and to localize the outlet 95, 96 in the model.

[0204] As an example, the robot 501 comprises different power cables, each configured to be compatible with a different power outlet. The robot is configured to classify the power outlets 95, 96 in the model based on their compatibility with the power cables. In particular, the robot 501 is configured to prioritize the power outlets based on their charging rate characteristics, e.g. to prefer the power outlet with the highest maximum charging rate.

[0205] The robot 501 can preferably charge at a charging station. However, e.g. based on a threshold criterion of the remaining energy level 97 of the robot 501, by evaluating the energy consumption to reach the charging station and the energy consumption to reach one of the locations of the power outlets 95, 96, the robot 501 is dispatched to one of the locations of the power outlets 95, 96.

[0206] Figure 12 Another embodiment of a robot 601 according to the invention is depicted, wherein the robot 601 comprises a power cable 93 and a power outlet / inlet that is compatible with a power inlet / outlet of another robot 98. The power cable 93 enables a connection between and a transfer of electrical energy between the two robots 601, 98, e.g. to the battery compartment 20 of the robot 601.

[0207] As an example, the robot 601 has a remaining power level 97 such that the robot 601 cannot reach a charging station nor a power outlet. Another robot 98 scheduled has a remaining power level 100 that allows power transfer to the robot 601. The robot 601 is configured to connect autonomously to the other robot 98 by connecting the power cable 93 to the power outlet 99 of the other robot 98.

[0208] Figure 13 An exemplary robot drive unit 110 for use in a metrology robot, such as a Figures 6 to 8 humanoid robot, with a high-precision measuring instrument, e.g. a metrology laser scanning unit, is illustrated in a 3D cross-sectional view. The drive unit 110 comprises inside a housing 111 an electric motor with a rotor 112 and a stator (not shown). For transferring rotor movement, the drive unit 110 further comprises a gearbox 113, e.g. a cycloidal gear as known in principle in the art. The gearbox 113 picks up rotor rotation at its gearbox input side or primary side I (on the left side in the figure) and transfers this rotation to its gearbox output side or secondary side O (on the right side in the figure). A gear output shaft or flange 114 is arranged at the gearbox output side O.

[0209] The drive module 110 further comprises an angle or rotary encoder 116 with a scale 118 and a sensing portion 117 for measuring the rotation or rotational position of the gearbox 113 and / or for measuring the position of the output axis 119. In contrast to known implementations, the rotary encoder 116 is arranged on the input side I of the drive unit 110 or (resp.) of the gearbox 113 (i.e. on the left side in the figure).

[0210] Hence, the encoder 116 is connected to the output side O by a connection 115. In this example, the connection 115 is embodied as a shaft in the center of the drive unit 110 that connects the flange 114 with the scale 118. The scale 118 arranged on the inner distal end of the connection shaft 115 is for example embodied as a glass ring or circle as part of an optical angle encoder 116, wherein the plane of the ring or circle is perpendicular to the rotation or output axis 119. Alternative measurement principles can for example be magnetic or capacitive principles as known in principle in the art. As shown, the glass ring 118 is rigidly attached to the connection shaft 115 by a shaft extension or shaft disc 120 that extends in parallel to the glass circle 118 at the input side I.

[0211] Accordingly, the output rotation of the gear 113 is transmitted or returned to the input side I through the connecting shaft 115, and the encoder scale 118 rotates identically with the gear output. The axis 119 passing through the motor shaft connects the gear output with the encoder disk 118, thereby bringing the output rotation to the high-precision encoder 116. Accordingly, by reading or sensing the scale 118 with the sensor device 117 of the encoder, the gear output rotation can be measured.

[0212] As an alternative to this description, the shaft disk 120 itself can be embodied as an encoder carrier or encoder rotor disk or it can comprise scale features (e.g. optical scale elements) that can be read or sensed by the encoder sensor 117. As a further example, one or more magnets are attached to the inner end of the shaft 115 to be sensed by a Hall sensor located at the PCB, which is an encoder device allowing for absolute rotation angle measurement.

[0213] As shown in the figures, the sensing part of the encoder or the active part of the encoder 116 is arranged on the left inner side of the housing 111, respectively. For example, in case of an optoelectronic measurement principle, light emitting and receiving units are arranged on the PCB for emitting light towards the glass ring 118 and receiving light reflected back from the encoding elements of the ring 118.

[0214] With the shown configuration, the encoder glass ring 118 can be mounted directly above the motor PCB. Accordingly, this arrangement allows to place the encoder sensor 117 on the same PCB as the motor electronics. Thus, a single PCB is sufficient and the drive unit 110 has a compact, space-minimized design.

[0215] The motor controller of the robot drive unit 110 can be programmed to use the position information from the rotary encoder 116 for motor commutation control. Accordingly, this design has only one common PCB, i.e. the encoder electronics are integrated on the motor circuit board, which is advantageous for such motor control arrangements using the output of the integrated rotary encoder. Additionally or alternatively, the motor control by rotary position information can be realized by an additional (low-accuracy) rotary sensor located at the motor rotor side (i.e. before the gear 113) as it is known in principle in the art. For example, a single Hall sensor (e.g. arranged as described above) using magnetic feedback of only one "coil interval" (360° / n coils) and / or of the motor coils for motor control.

[0216] Figure 14 An embodiment of a robot drive joint as a dual-axis drive joint 121 as can be used in a robot according to the present application is shown in a 3D cross-sectional view (top plate) and a 2D cross-sectional view (bottom plate).

[0217] WithFigure 13 The example depicted in the middle is similar, the drive unit 121 comprises a motor (here denoted external motor) with an (external) stator (coil) 122 and an (external) rotor 112, in this example the rotation of the motor is transmitted through a gear 113, such as a cycloidal gearbox, to a gear output 114, for example for driving a first robot part. As explained in more detail above, the rotation on the gear output side is consistently transmitted via a rigid connection 115 to the inside, to a (first) encoder 116, or more precisely to a (first) scale 118, such as a glass circle scale. The first scale 118 can be sensed by a sensing part 117 of the first encoder 116, which is arranged at a PCB 123 in the interior of the drive unit 1, for example integrated in the motor circuit board.

[0218] Above the first or external motor arrangement with the gear 113 and the encoder 116 for reading the gear output rotation on the gear input side, the drive unit 121 comprises a second or internal motor arrangement, whereby the drive unit is a dual shaft drive unit 121. This internal motor comprises an internal stator 131 and an internal rotor 132, which leads to an output side of the drive unit 121. However, in contrast to the external drive without a gear transmission, here a direct internal motor output 134 is provided, for example for driving a second robot part.

[0219] The drive unit 121 further comprises a second encoder 136 for measuring the rotation of the internal motor or internal rotor. This “internal” encoder 136 also comprises a scale 138, for example a glass circle or other form of encoding carrier known in the art. The scale 138 is read by a sensor 137. In the same way as the first sensor 117, the second sensor 137 is arranged on the other side of the PCB 123, thus, the drive unit 121 comprises a “central” PCB, which can also comprise a motor circuit with encoder sensors 117, 137 on both sides. Thus, only one static PCB 123 and only one microcontroller are needed for all motors and for the encoders of the two independent motors or drives.

[0220] The drive unit 121 further comprises a static electrical and mechanical interface 130 with an electrical connector or a magnetic connector, which can for example be embodied for attachment with a snap or with a screw to a hollow center shaft 140, for example for cable feed therethrough, and which can comprise an integrated cooling indicated schematically by a gill 139. Thus, a dual shaft drive unit 121 is provided, with a common housing and interface 130 for two drives.

[0221] This dual shaft drive unit 121 with or without gearbox 113 shows a relatively simple static interface (regarding power cable routing and mechanics) and further provides the advantage that the last movable element can be operated completely 360° without any cable routing limitations. Both drives can rotate uninterrupted around the rotational axis 119 a full revolution with independent speed and direction from each other. Thereby, each output is sensor controlled regarding angle, torque and / or force by corresponding angle sensors 116, 136, respectively, and thus controlled.

[0222] The dual shaft drive module 121 can be modified or show additional features or appropriate additional components as mentioned for the embodiments described, e.g. an additional (low accuracy) rotation sensor at the motor rotor side. Figure 13

[0223] Although the present application has been illustrated above partly with reference to some specific embodiments, it must be understood that numerous modifications and combinations of different features of the embodiments are possible. All such modifications are within the scope of the appended claims.​

Claims

1. A mobile robot, the mobile robot comprising: • a locomotion unit configured to provide locomotion of the robot over a ground surface, • a mechanically actuated multi-joint articulation system comprising a plurality of joints and configured to provide movement of an interaction component relative to a reference point on the robot, • the interaction component comprising a probing sensor configured to provide optical and / or tactile distance probing of a surface of an object to be measured by the robot, and • a position determination device configured to provide determination of angular positions of the plurality of joints, characterized in that • the robot comprises a docking interface, wherein the docking interface is configured to provide docking of a docking point of the robot to a docking counterpart device, wherein the interaction component is connected to the docking point via a subset of the plurality of joints such that a degree of freedom of movement of the interaction component relative to the docking point depends on joint positions of the subset of the plurality of joints and is independent of joint positions of a remainder of the plurality of joints, • the position determination device is configured to provide increased accuracy of determination of angular positions for the subset of the plurality of joints compared to the remainder of the plurality of joints, and • the robot is configured to provide a 3D scan by moving the subset of the plurality of joints and by the probing sensor taking measurements.

2. The mobile robot of claim 1, wherein, The docking is a releasable mechanical docking, and in a docked state the interaction component is mechanically connected to the docking counterpart device via the subset of the plurality of joints and the docking interface.

3. The mobile robot of one of claims 1 to 2, wherein, The docking interface is configured to provide the docking as a rigid docking with the docking counterpart device.

4. The mobile robot of claim 3, wherein, The rigid docking is provided by a three-point support between the docking interface and the docking counterpart device.

5. The mobile robot of one of claims 1 to 4, wherein, The docking interface is configured to provide that all six degrees of freedom in space are fixed when the docking interface is docked to the docking counterpart device.

6. The mobile robot of one of claims 1 to 5, wherein, The docking interface comprises a docking sensor unit configured to continuously determine a 6DoF change of position relative to the docking counterpart device.

7. The mobile robot of claim 6, wherein, The docking sensor unit comprises a capacitive sensor with a measurement range of less than 1 cm.

8. The mobile robot of one of claims 6 to 7, wherein, The docking sensor unit comprises a video camera.

9. The mobile robot of one of claims 1 to 8, wherein, Each joint of the subset of the plurality of joints comprises a robot drive module for driving rotational joint movement, wherein the robot drive module comprises: • a rotational drive comprising a motor circuit board, a stator and a rotor, wherein the rotor is configured to be controlled by the motor circuit board to rotate relative to the stator about a rotational axis, • a gearbox configured to convert rotational movement of the rotor about the rotational axis into rotational movement of a gearbox output component about the rotational axis according to a defined gear ratio, wherein the motor circuit board and the stator are arranged axially on one side of the gearbox relative to the rotational axis, is denoted a gearbox input side, and the gearbox output member engages the gearbox from the other side of the gearbox denoted a gearbox output side, and • a rotary encoder configured to detect rotation of the gearbox output member about the rotation axis, wherein • the drive module comprises a connection portion extending from the gearbox output side to the gearbox input side and configured to pick up the rotation of the gearbox output member in a rigid manner, thereby providing rotation of the connection portion in line with the rotation of the gearbox output member, i.e. the same rotation of the connection portion as the rotation of the gearbox output member; and • the rotary encoder is arranged on the gearbox input side and configured for measuring rotation of the connection portion about the rotation axis.

10. The mobile robot of one of claims 1 to 9, wherein, The robot comprises a laser tracker configured to be arranged in a fixed positional relationship with the docking point, wherein the laser tracker is configured to provide automatic laser-based tracking of a portion of the robot moved by at least one joint of the subset of the plurality of joints, and to determine 3D positional data of the portion of the robot.

11. The mobile robot of one of claims 1 to 10, wherein, The robot comprises a camera arrangement configured to be arranged in a fixed positional relationship with the docking point, wherein the camera arrangement is configured to provide automatic image-based tracking of another portion of the robot moved by at least one joint of the subset of the plurality of joints, and to determine 3D positional data of the other portion of the robot, in particular wherein the other portion comprises a visual pattern for assisting the image-based tracking.

12. The mobile robot according to one of claims 1 to 11, wherein • the robot comprises a further interaction member configured to grasp objects and to move by moving a plurality of joints of the multi-jointed articulation, wherein the further interaction member is connected to the docking point via a further subset of the plurality of joints, such that a degree of freedom of movement of the further interaction member relative to the docking point depends on joint positions of the further subset of the plurality of joints and is independent of joint positions of the subset of the plurality of joints and of a remainder of the plurality of joints, • the position determination arrangement is configured to provide an increased angular position determination accuracy for the further subset of the plurality of joints compared to the remainder of the plurality of joints, e.g. similar to the position determination accuracy of the subset of the plurality of joints, and • the robot is configured to provide the 3D scan by grasping an object to be measured by the further interaction member and moving the further subset of the plurality of joints.

13. The mobile robot of one of claims 1 to 12, configured to derive pose information of the coupling partner device relative to a mounting platform carrying the coupling interface, wherein, The coupling pairing arrangement is arranged on the mounting platform and is configured to interact with the coupling interface to provide the docking.

14. The mobile robot of claim 13, wherein, The robot is configured to access the pose information by electronic communication through the docking point.

15. The mobile robot of one of claims 13 to 14, wherein, The robot comprises an optical perception sensor and is configured to • performing pattern recognition and pattern analysis of a pattern arranged on the coupling counterpart and / or the mounting platform using the perception sensor, and • associating the pattern with a pose parameter providing the pose information, in particular by accessing a database comprising different patterns and associated pose parameters for individual ones of the different patterns.

16. The mobile robot of one of claims 1 to 15, wherein, The coupling counterpart is arranged at a measurement platform configured to support an object to be measured, and the measurement platform comprises a first coupling counterpart and a second coupling counterpart, wherein the first coupling counterpart and the second coupling counterpart are arranged spaced apart from each other and each of the first coupling counterpart and the second coupling counterpart is configured to interact with the coupling interface to provide a docking of the docking point of the robot to the measurement platform, wherein the robot is configured to: • derive first pose information of the first coupling counterpart relative to the measurement platform and derive second pose information of the second coupling counterpart relative to the measurement platform; and • provide 3D scan data for the object to be measured by: o docking to the first coupling counterpart and providing first distance probe data for a first docking position relative to the first coupling counterpart by distance probing using the probe sensor, o docking to the second coupling counterpart and providing second distance probe data for a second docking position relative to the second coupling counterpart by distance probing using the probe sensor, and o merging the first distance probe data and the second distance probe data using the first pose information and the second pose information.

17. The mobile robot of claim 16, wherein, The robot is configured to perform a docking pose adjustment measurement, wherein the docking pose adjustment measurement comprises: • selecting a set of essential features, wherein o the first distance probe data comprises a pose of a feature of the set of essential features relative to the first coupling counterpart, and o the probe sensor has access to the feature of the set of essential features when the robot is docked to the second coupling counterpart, • comparing poses of features of the set of essential features in the first distance probe data and the second distance probe data, and • merging the first distance probe data and the second distance probe data using the comparison. • selecting a set of essential features, wherein o the first distance probe data comprises a pose of a feature of the set of essential features relative to the first coupling counterpart, and o the probe sensor has access to the feature of the set of essential features when the robot is docked to the second coupling counterpart, • comparing poses of features of the set of essential features in the first distance probe data and the second distance probe data, and • merging the first distance probe data and the second distance probe data using the comparison.

Images