EP4096552A1

EP4096552A1 - Machine and control system of a robotic device

Info

Publication number: EP4096552A1
Application number: EP21707370.9A
Authority: EP
Inventors: Loredana ZOLLO; Clemente LAURETTI; Francesca CORDELLA
Original assignee: Masmec SpA; Universita' Campus Bio-Medico di Roma (UCBM)
Current assignee: Masmec SpA; Universita' Campus Bio-Medico di Roma (UCBM)
Priority date: 2020-01-31
Filing date: 2021-02-01
Publication date: 2022-12-07
Also published as: WO2021152556A1; IT202000001900A1

Abstract

A control method for a robotic manipulator cooperating with the surgeon, in particular for application in spine surgery, specifically in a tapping operation of the vertebral peduncles. The general aim of the invention is to guarantee to the surgeon to have a continuous control of the force/torque applied on the vertebral peduncles whereas the robotic arm constraints the position and orientation of the surgical instrument along the axis of motion identified in the pre-operatory phase. The method provides that the surgeon controls directly and continuously the interaction forces/torques along the tapping axis, whereas the robotic arm constraints the position and orientation of the surgical instrument in the directions orthogonal to the tapping axis identified in pre-operative phase.

Description

MACHINE AND CONTROL SYSTEM OF A ROBOTIC DEVICE

DESCRIPTION Technical field of the invention

The present invention relates to an apparatus and to a control system associated to robotic manipulator, the latter in particular for spine surgery. Background

The insertion of peduncle screws has become a standard procedure in the clinical practice, especially for the stabilization of the lumbar or thoracic tract of the spine in patients with spinal instability or deformity.

Due to anatomical proximity of the vertebral peduncles with delicate tissues (such as the spinal cord, the roots of nerves and blood vessels), it is necessary that the positioning of the screws is accurate, indeed to avoid neurological, vascular and visceral damages to the patient.

The so-called tapping procedure consists in threading the patient’s peduncle of for a more accurate insertion of the screws, with the purpose of reducing significantly the risk of malpositioning the screw itself.

In the last decade, the adoption of robotic devices in the spine surgery has increased the repeatability and accuracy of operations for fixing the spine. The robots in fact can reach upper precision levels, if one compares a robot- mediated operation to the manual operations performed by the surgeon, they are not affected by fatigues and tremors and they carry out repetitive activities without reducing performances.

The literature proposes different robotic systems for tapping or drilling of the screw path. Such systems can be divided into three main categories: i) teleoperated robots, typically wholly remotely guided by the surgeon through a haptic interface; ii) passive robots, which provide a guide to the surgeon during the drilling phase, the latter performed wholly manually; and iii) active robots which, based upon pre-planned trajectories, perform the drilling procedure of the peduncle autonomously under the complete control of a computer. An example of teleoperated robot for spine surgery is the Cooperative Robotic Assistant (CoRA) [J. Lee, I. Hwang, K. Kim, S. Choi, W. Kyun Chung, Y. S. Kim - Cooperative robotic assistant with drill-by-wire end-effector for spinal fusion surgery - Industrial Robot: An International Journal, 36(1): 60-72, 2009], which is based upon a master/slave control system mechanically decoupled to perform a high-speed drilling of the peduncle. In other words, the surgeon grasps and rotates a handpiece which is connected to a motor integral with the end member of the master-side robot. The handpiece rotation angle is measured through position sensors integrated in the motor itself and it is converted, through a ball screw mechanism, into the linear motion of a high speed drill, assembled on the end member of the slave-side robot.

Such system has the double advantage of increasing the accuracy of the surgical operation and of reducing the surgeon’s muscle fatigue with respect to a manual spine surgery operation. On the other hand, the surgeon does not control finely the rotation speed or the torque of the surgical instrument, to the detriment of accuracy and reliability of the whole operation.

Moreover, this system does not allow the surgeon to perform tapping of the patient’s peduncle in assisted or automatic way. A significant physical and cognitive effort is then requested to the surgeon in order to complete such surgical procedure, running the risk of adopting incorrect postures and incurring in musculoskeletal damages.

An example of passive robot is the commercial device MazorX[ A. Ghasem, A. Sharma, D. N. Greif, M. Alam, M. Al Maaieh - The arrival of robotics in spine surgery: a review of the literature - Spine, 43(23): 1670-1677, 2018] Such robot is used as device for positioning the surgical instrument and provides a passive mechanical guide during use. Therefore, the orientation and position of the surgical instrument (the drill) are constrained by the robotic device and the surgeon carries out manually the peduncle tapping.

This system, like the previous one, does not allow the surgeon to perform tapping of the patient’s peduncle in assisted or automatic way.

An example of active robot which can perform an operation of spine surgery wholly autonomously is Spinebot [G. B. Chung, S. Kim, S. G. Lee, B.-J. Yi, W. Kim, S. Min Oh, Y. S. Kim, B. R. So, J. II Park, S. H. Oh - An image-guided robotic surgery system for spinal fusion - International Journal of Control, Automation, and Systems, 4(1): 30-41, 2006] A module of such robot plans the trajectories which have to be followed, by keeping into consideration even the unvoluntary motions of the patient, thanks to the use of optoelectronic systems and markers placed on the patient and on the surgical instrument. Such trajectories then are performed by the robot thanks to a position control of PID (Proportional-Integral-Derivative) type.

This system has the advantage of not exposing the surgeon to incorrect positions and not to subject him/her to muscle fatigue during the operation, since the drilling is performed by the robot in wholly autonomously.

However, one of the main disadvantages of such robotic device is lack in supervision by the surgeon on the forces and/or torques which are exerted on the peduncle by the surgical instrument.

The literature then proposes even robotic systems which use semi-autonomous control methods for tapping or drilling the screw path. Such methods allow the surgeon to control directly the position and speed of the surgical instrument but not the forces and/or torques which are exerted on the peduncle. An example of semi-autonomous control method is shown US2019/0090966A1 , in which based upon an implementation of the proposed system, the surgeon applies a force on the surgical instrument and such force is translated into linear advancement of the same instrument along the tapping axis (linear speed proportional to the force applied by the surgeon); such linear advancement then generates a rotation of the surgical instrument around the tapping axis with angular speed proportional to the linear advancement speed of the instrument and to the screw pitch. It follows that the surgeon, by modulating the force applied on the surgical instrument, is capable of controlling directly not only the linear and rotational tipping speed; the torque applied by the surgical instrument is not managed directly by the surgeon but, on the contrary, is controlled by an automatic mechanism which monitors it and compares it with a threshold. When the rotation torque exceeds the prefixed threshold, the instrument rotation is interrupted.

Such approach, which then allows the surgeon to manage exclusively the motion speed of the surgical instrument and not to control directly and actively the force and the torque, involves the risk of damaging tissues in contact with the surgical instrument. Summary of the invention

The technical problem placed and solved by the present invention is then to provide a robotic apparatus and a related control method allowing to obviate the drawbacks mentioned above with reference to the known art.

Such problem is solved by an apparatus according to claim 1 and by a method according to claim 13.

Preferred features of the present invention are set forth in the depending claims.

The invention provides an apparatus based upon a surgeon-robot shared control method, and in particular a robotic manipulator configured for spine surgery, more specifically for a tapping procedure of a vertebral peduncle. The method allows the surgeon to perform tapping semi-autonomously by allowing him/her to control the torque and the force applied according to the tapping axis and by constraining automatically, instead, at least the spatial orientation of such axis.

In particular, according to preferred embodiments the surgeon can control the end member (“end effector") of the robotic manipulator, thereto a tapping tool is connected, along an axis of motion identified in pre-surgery phase and corresponding to the above-mentioned tapping axis. The surgeon manages continuously the forces and torques applied to the patient’s spine according to the tapping axis, whereas the position and orientation of the end member of the manipulator are constrained along the directions orthogonal to said axis, wherein in fact a high positioning accuracy is requested. Therefore, the robotic manipulator is wholly passive along the tapping axis and, in embodiment variants, it can be moved along such axis by the surgeon who exerts controlled forces and/or torques.

Advantageously, the surgeon interacts with the manipulator by means of an approach of so-called “ hands-on " type, that is by manoeuvering directly a portion of the manipulator or a remote unit in communication with the manipulator itself.

Based upon a preferred embodiment, the control method provides that, during the peduncle tapping, the torque exerted by the robotic arm around the tapping axis is directly proportional to the force exerted by the surgeon along the same axis.

A control variant instead can provide that, during the peduncle tapping, the force exerted by the robotic arm along the tapping axis is directly proportional to the torque exerted by the surgeon around the same axis.

The invention then guarantees to the surgeon to keep a direct and continuous control of the interaction force and/or torque applied on the vertebral peduncle along the tapping axis, without losing accuracy in position along the other directions.

The invention further allows to reduce significantly the surgeon’s physical and cognitive effort and the relative fatigue and reduces the risk of incorrect postures and musculoskeletal damages. Both operator ergonomics and safety for the patient then result to be maximized.

The same approach proposed herein in the context of an apparatus and a method for spine surgery can be used even in different medical fields, for example orthopaedic surgery. Moreover, the proposed apparatus and method are suitable for not medical applications, such as industrial robotics, for example for the assembly of components, and all fields in which it is possible to facilitate the execution of operations which currently require a high cognitive and physical effort by the operators, in addition to high accuracy. The proposed apparatus can be used to perform indifferently procedures of drilling, threading and screwing/unscrewing bolts, nuts or screws. In general terms, the invention can be easily integrated in all manipulators with at least 4 levels of freedom and it can be used even for teleoperation of industrial manipulators operating in environments hostile or dangerous for man, for example radioactive, underwater environments or, more generally, environments having a chemical, physical or biological risk for man.

Other advantages, features and use modes of the present invention will result evident from the following detailed description of some embodiments, shown by way of example and not for limitative purposes.

Brief description of figures The figures of the enclosed drawings will be referred to, wherein:

^■ Figure 1 shows a block diagram exemplifying the general structure and control architecture of a robotic apparatus according to a preferred embodiment of the present invention, which provides a control of the torque exerted by a relative surgical instrument, or end member, around the tapping axis;

^■ Figure 2 shows another block diagram exemplifying the general structure and control architecture of a robotic apparatus according to a preferred embodiment of the present invention, which provides a control of the rotation speed of the surgical instrument around the tapping axis;

^■ Figure 3 shows another block diagram exemplifying the general structure and control architecture of a robotic apparatus according to a preferred embodiment of the present invention, which provides a control of the force which the surgical instrument applies along the tapping axis; ^■ Figure 4 shows another block diagram exemplifying the general structure and control architecture of a robotic apparatus according to a preferred embodiment of the present invention, which provides a control of the linear speed of the surgical instrument along the tapping axis;

^■ Figure 5 shows a representation exemplifying a surgeon-robot configuration of the apparatus of Figure 1 or 2, highlighting the positioning of reference cartesian triads;

^■ Figure 6 shows a flowchart of an embodiment of a control scheme for the apparatus of Figure 1 or 2;

^■ Figure 7 shows a diagram in more detail of an embodiment of surgeon-robot shared control scheme for the apparatus of Figure 1 or 2;

^■ Figure 8 shows another diagram in more detail of an embodiment of surgeon-robot shared control scheme, which is decoupled on the three axes of a reference triad of an end effector of the apparatus of Figure 1 or 2;

^■ Figure 9 shows a representation exemplifying a surgeon-robot configuration of the apparatus of Figure 1 or 2, with control action in force on an axis z of a reference triad of the surgical instrument corresponding to a tapping axis; and

^■ Figure 10 shows a representation exemplifying a surgeon-robot configuration of the apparatus of Figure 1 or 2, with control action in speed on an axis z of a reference triad of the surgical instrument corresponding to a tapping axis. Detailed description of preferred embodiments

Hereinafter various embodiments and variants of the invention will be described, and this with reference to the above-mentioned figures. Analogous or corresponding components are meant designated, in the several figures, with the same numeral or alphabetic reference.

In the following detailed description, additional embodiments and variants with respect to embodiments and variants already treated in the same description will be illustrated limitedly to the differences with respect to what already illustrated.

Moreover, the several embodiments and variants described hereinafter are likely to be used in combination, where compatible.

By firstly referring to Figure 1, a robotic apparatus according to a preferred embodiment of the invention is designated as a whole with 100.

The apparatus 100 includes a robotic manipulator, or arm 1, provided with end member 11 (“end effector") arranged at a distal end of the manipulator 1 itself. The end member 11 - which is, or bears integrally, a surgical instrument - is provided with a working axis A according thereto the end member, or a distal portion thereof, applies a force and/or a torque on a body district.

In the present example, the end member 11 is, or bears integrally, a tapping tool, or instrument, configured for threading a peduncular seat of a vertebra according to the working axis A, which then is a tapping axis. The robotic manipulator 1 includes a base 12 configured for remaining fixed with the operating environment and a mobile portion 13, which includes the end member 11 and which further comprises a maneuvering portion 14 configured indeed for being manoeuvred by an operator, in particular a vertebral surgeon, according to the above-mentioned approach of “hands on" hand. Embodiment variants can provide that the operator performs a manipulatory control on a remote unit of the apparatus 100 and that such manipulatory action is transferred, according to the control method described hereinafter, to the end member 11. The apparatus 100 further includes a control unit 101 configured for controlling the manipulator 1 , with particular reference to the linear and/or angular position of the end member 11 along/according to the axis A and/or to the force and/or torque applied according to such axis. The control unit 101 further constrains the linear and angular position of the end member 11 in the directions orthogonal to the axis A.

The control unit 101 can be incorporated in the manipulator 1 or arranged, wholly or partially, at a remote station of the apparatus 100 and in communication with control elements integrated in the manipulator 1 itself. The apparatus 100 further includes force sensors or transducers 21 and/or torque sensors or transducers 22, to detect the force and/or the torque exerted by the operator according to the tapping axis A and/or by the end member 11 on the body district.

The apparatus 100 further includes linear and/or angular position sensors or transducers 23 of the end member 11 and/or of the maneuvering portion 14 actuated by the surgeon. A portion of such sensors or transducers 23 can even be arranged on the body district, in particular on a vertebra of the patient. Advantageously, such sensors or transducers 23 are portion of, or cooperate with, a viewing system 30. The latter in particular consists of optoelectronic means and, in such embodiment, the sensors or transducers 23 can be, or include, optical markers.

Based upon the data provided by the viewing system 30 in real time and/or of pre-stored data inherent the anatomy of the operated body district - for example X-rays of the peduncle performed in pre-surgery phase - the control unit 101 can receive or determine a wished position for the end member 11, it can perform a comparison with the current position and consequently control the end member 11 itself.

The sensors or transducers 21, 22 and 23 and components associated thereto can have different collocation on the manipulator 1 , on portions thereof or on the operated subject and, even for this reason, they are represented in Figure 1 functionally, without showing the specific collocation thereof.

A simplified embodiment variant can provide that the tapping axis A has constrained, fixed and predetermined position.

Based upon the control scheme represented schematically in Figure 1 and 3, the force and/or the torque measured by the sensors or transducers 21 and/or 22 are weighed, one or both, through a scale factor determined empirically and then provided as input, as independent values or values associated through a mathematical operation or equation, for example of differential type, to a force control module of the control unit 101.

Similarly, in the herein considered example, the wished position and the measured position of the end member 11 are provided as input, as independent values or values associated through a mathematical operation or equation for example of differential type, to a position control module of the control unit.

Through an additional control module based, for example, on an inverse dynamic algorithm, the (for example current) control signal for each actuator of the robot is determined as consequence from the force and/or torque exerted by the operator in the maneuvering area 14. Equally, the position and orientation of the end member 11 are imposed according to the planes orthogonal to the axis A.

Figures 2 and 4 show a scheme analogous to the one of Figures 1 and 3, wherein the measured torque and force are replaced, respectively, by an angular speed and a linear speed, measured through sensors or transducers 24 and then the control unit 101 includes a speed control module.

Based upon this preferred control variant, the operator finely controls the rotation speed of the surgical instrument 11 around the tapping axis A by simply modulating the force which he/she exerts along the tapping axis A.

In all described control modes, the apparatus 100 is configured so that the linear and angular position of the end member 11 with respect to a plane orthogonal thereto, and in particular the one of an axis thereof corresponding to the tapping axis A, is determined, that is imposed, by the control unit 101. According to a shared control mode which is one of the salient aspects of the invention, the operator action instead determines one or more of the following quantities associated to the action of the end member 11 according to the axis A: applied force, applied torque, angular and linear speed and angular and linear position. In other words, the position and orientation of - and preferably the stresses applied by - the end member 11 in the directions orthogonal to the tapping axis A are constrained by the command of the control unit 101 and then they cannot be modified by the operator who acts on the maneuvering portion 14 of the instrument.

On the contrary, the control action in position, speed, torque and/or force according to the tapping axis is passive or exerted depending upon the operator inputs. Based upon a preferred embodiment, the torque control action is configured so that the torque exerted by the end member 11 around the tapping axis A is directly proportional to the force exerted by the surgeon along the same axis, the latter advantageously measured through the sensors or transducers 21.

According to a different embodiment, the force control action is configured so that the force exerted by the end member 11 around the tapping axis A is directly proportional to the torque exerted by the surgeon on a maneuvering portion (for example rotating handpiece mechanically coupled with the tapper), the latter advantageously measured through the sensors or transducers 21. The advantage of being able to apply a torque on a maneuvering portion, apart from a force along the tapping axis, allows to modulate the sensitivity therewith the surgeon controls the insertion of the tapper inside the peduncle by simply variating the diameter of the rotating handpiece. The more the diameter of the rotating handpiece is, the greater is the sensitivity therewith the surgeon controls the insertion of the tapper in the peduncle. The fact of controlling the insertion of the tapper inside the peduncle with the right sensitivity allows the surgeon to reach in a finer and more precise way the tapping end point (established by pre-planned trajectories which are shown to the surgeon on a monitor together with the position of the surgical instrument measured by the optoelectronic system). This decreases the probability that the underneath tissues could be damaged by the tapper and further increases the operation safety.

Such strategy then guarantees to the surgeon to control directly the interaction forces/torques along the tapping axis A, without losing accuracy in the positioning of the surgical instrument in the directions orthogonal to the tapping axis. With reference to Figures 5 and 6, the dynamic model of a robotic device at n Degree of Freedom (DF) is described by the following equations: wherein B(q) is the inertia matrix, is the matrix describing the centrifugal and Coriolis effects, F_v is the viscous friction torque, is the static friction torque, g(q) is the gravity contribution, h_s, is the vector of forces and moments exerted by the end member of the robot on the environment, is the transposition of the geometrical Jacobian, are respectively the position, speed and angular acceleration and τ is the torque at the joints delivered by the robot actuators. Given the equation of the dynamic model (1), the control strategy implemented in the apparatus 100, in particular to determine the torques at the joints which have to be delivered by the robot actuators, can be derived by an inverse dynamics control expressed by the following equation wherein t_e is the control toraue enterina the actuators_, is an estimate of the inertia matrix of the robot, is an estimate of the matrix describing the centrifugal and Coriolis effects is an estimate of the viscous friction torque, is an estimate of the static friction torque and is an estimate of the gravity contribution. It is important to specify that the control strategy implemented in the apparatus 100 can be derived by any conventional control of centralized type, for example a control with gravity compensation. However, an inverse dynamics control guarantees higher performances in terms of surgeon-robot interaction.

Based upon a preferred embodiment of the present invention, the conventional control law is formulated as follows: wherein is the right pseudo-inverse matrix of the geometric or analytical Jacobian (this choice depends upon the representation that is used to define the orientation) and are, respectively, the error in speed and position expressed in the basic reference triad

The vectors are respectively the Cartesian position and speed error of the end member 11 of the manipulator 1. On the contrary, are respectively the position and angular velocity of the same.

With the purpose of meeting the accuracy requirements requested by the tapping procedure, as already mentioned the reference laying of the end member 11 of the manipulator 1 can be pre-planned through a system of surgical navigation which, depending upon the position of the peduncle to be threaded, detects the optimum direction of the tapping axis A. Thanks, for example, to the use of the optoelectronic system 30 with markers suitably positioned on a vertebra of the patient and on the end member 11 , the optimum direction of the tapping axis A can be detected, in real time, with respect to the position of the peduncle to be threaded.

In other words, a reference triad [X_cam, Y_cam, Z_cam ] defined so that Z_cam coincides with the versor of the tapping axis e X_cam and Y_cam are selected so that the triad [X_cam, Y_cam, Z_cam ] is orthogonal and right-handed. Under the ideal condition, in which the patient is immobile, the wished laying of the end member 11 of the manipulator 1 remains constant and the speed and acceleration are null. On the contrary, under not ideal conditions in which the patient’s motions have to be taken into account during the surgical operation, the reference laying of the robot should be defined as follows

(5) wherein are respectively laying, speed e acceleration of the reference triad [X_cam, Y_cam, Z_cam ] with respect to the basic reference triad [X_B,Y_B,Z_B] A reference laying like the one shown in (5) guarantees that the control action in position contained in (4) constraints the axis z of the triad (Figure 3) to be coincident with the axis z_cam of the reference triad [X_cam, Y_cam, Z_cam ] that is the tapping optimum axis detected by the optoelectronic system.

It is interesting to note that, thanks to the use of the optoelectronic system 30 and associated markers, the unvoluntary motions of the patient, such as for example those due to breathing, can be correct and then not alter the relative position between the patient’s peduncle and the surgical instrument, thus allowing to keep a high accuracy during the procedure.

In the equation (4), Ad₁ is the transposition of the added matrix that has been suitably inserted in order to transform the position i and speed £ error from the basic reference triad [X_B,Y_B,Z_B] to the reference triad of the end member del robot [X_tool, Y_tool, Z_tool ] and it is expressed as follows wherein is the rotation matrix that expresses the orientation of the reference triad of the end member, [X_tool, Y_tool, Z_tool ] with respect to the basic one, [X_B,Y_B,Z_B]. Whereas is the symmetric skew matrix of the position of the end member, P_tool.

An important advantage introduced by the added matrix Ad_t is the possibility of decoupling the control law on the cartesian axes of the reference system of the end member [X_tool,Y_tool,Z_tool] (as shown in Figures 5 and 8). This allows to apply a control action in position in the directions in which a higher position accuracy is required (that is the axes X_tool e Y_tool) and a control action in force and/or torque in the directions in which a finer control of the interaction force/torque between the surgical instrument and the peduncle to be threaded is required. This can be obtained by selecting suitably the matrices of the gains of the control action in position, K_p and K_D, and force, K_p, as shown hereinafter.

K_p and K_B are the matrices of the gains of the position control and are 6 x 6 semi-defined positive diagonal matrices wherein are null so that the control action in position has no effect along and around the axis z (tapping axis). From here, the surgeon is capable of moving the end member of the robot along such axis, by means of the above-mentioned approach of “hands-on” type.

On the contrary, K_F is an identity matrix 6 x 6 and x_f - c_F(F_d — F_e) is the control action in force/torque which has been specifically devised so as to allow the surgeon to control the torque and the force exerted by the surgical instrument on the spine of the patient.

Figure 5 shows a block diagram of the control action in force/torque acting on the tapping axis.

C_F is the matrix of the gains 6 X 6 of the force control defined as

F_d and F_e are respectively the wished and measured forces expressed in the reference triad of the end effector of the robot

F_e is the force exerted by the surgeon on the patient along the direction of the tapping axis, which is filtered and scaled by a factor c.

The robot-patient interaction force, F_e, can be measured directly, by means of a sensor positioned on the tip of the surgical instrument or by means of torque sensors integrated in the robot by exploiting the following equation, wherein are the torques measured by the robot sensors and is the inverse of the transposed Jacobian.

From the equations (9)-(11) and from the block diagram shown in Figures 7 and 9 it is evident that the torque control action was devised so that the torque exerted by the tapper around the tapping axis, T_ez, is finely modulated by the surgeon which exerts controlled forces, F_ez, along the direction of the tapping axis.

A variant of the control action in torque, proposed in Figure 9, is the control action in angular speed shown in Figure 10 and described by the following equation wherein ω_ez is the angular speed of the surgical instrument (the tapper) around the tapping axis and are the gains of the control in speed. Such gains are suitably selected with the purpose of allowing the surgeon to control finely the rotation speed of the surgical instrument around the tapping axis by simply modulating the force, F_ez, which he/she exerts on the patient along the tapping axis.

In the same way, other variants of the control action are described by the following equations (14) describes the control action in force devised so that the force exerted by the tapper along the tapping axis, F_ez, is finely modulated by the surgeon who exerts a controlled torque, T_ss, around the tapping axis.

(15) describes the control action in linear speed devised so that the linear speed therewith the tapper moves along the tapping axis, V_ez , is finely modulated by the surgeon who exerts a controlled torque, T_ez, around the tapping axis.

It will be understood that the above-illustrated control method can be used in the context of an operation for fixing, that is stabilizing, the spine, which typically is divided into four phases, that is: (a) decortication of the peduncular dorsal cortex (by means of an awl); (b) mapping and/or drilling of the path of the peduncle screws, aimed at obtaining a thread or a simple hole, which can be performed with the apparatus and the sofar described method; (c) positioning of the screws; and (d) connection of the screws to a rod which opens out the vertebrae, in case the intervention is aimed at reducing compression on the local nervous structures.

The present invention has been sofar described with reference to preferred embodiments. It is to be meant that other embodiments belonging to the same inventive core may exist, as defined by the protective scope of the herebelow reported claims.

Claims

1. A robotic apparatus (100) configured for the application of a torque/force to an external element, which robotic apparatus comprises:

- an end member (11) movable in translation and/or rotation according to a working axis (A) thereof;

- a maneuvering portion (14) associated to said end member (11) and operable by an operator;

- detecting means (21, 22, 24) configured to detect an operating quantity selected from the force, torque and linear or angular speed applied by the operator at said maneuvering portion (14);

- a control unit (101 ) of said end member (11 ), configured for:

^■ calculating, by means of a centralized control algorithm for example based upon inverse dynamics, an effective quantity, selected in a group comprising torque and force, according to said working axis (A) that said end member (11 ) must apply to the external element as a function of the quantity detected by said detecting means (21 , 22, 24), wherein the orientation and the Cartesian position of said end member (11) are constrained in the directions orthogonal to said working axis (A) and cannot be modified by the operator.

2. The apparatus (100) according to claim 1, wherein said control unit (101) is configured for calculating said effective quantity as proportional, directly or inversely, to said operating quantity.

3. The apparatus (100) according to claim 1 or 2, wherein said control unit (101) is configured to constrain said spatial orientation and position of said end member (11) in the directions orthogonal to said working axis (A) based upon detected position data and/or pre-stored position data.

4. The apparatus (100) according to any one of the preceding claims, wherein said end member (11) or a part of it is movable in translation along said working axis (A) and wherein preferably the corresponding axial displacement is determined by said control unit (101) as a function of a maneuvering action performed by the operator at said maneuvering portion (14).

5. The apparatus (100) according to any one of the preceding claims, wherein said detecting means comprises sensors and/or transducers of force and/or torque and/or angular or linear speed (21 , 22, 24).

6. The apparatus (100) according to any one of the preceding claims, comprising additional detecting means (23) configured to detect a linear and/or angular position of said end member (11 ) or of the external element.

7. The apparatus (100) according to any one of the preceding claims, comprising an optoelectronic system (30) for the detection and/or estimation of linear or angular position, speed or acceleration quantities.

8. The apparatus (100) according to any one of the preceding claims, wherein said control unit (101) comprises a module for correcting movements of the external element, in particular a body district of a patient.

9. The apparatus (100) according to any one of the preceding claims, wherein said control unit (101) is configured to decouple a control law exercised according to the cartesian axes of a reference system integral with said actuating device (11), in such a way to selectively apply a control action in position, in force and/or in torque on each axis.

10. The apparatus (100) according to any one of the preceding claims, which includes a robotic manipulator (1) and wherein said end member is an end tapping tool (11) configured to perform a threading of a vertebral peduncle, said working axis being a tapping axis (A).

11. The apparatus (100) according to the preceding claim, which comprises means for inserting a peduncle screw into a vertebral seat obtained by said tapping tool (11).

12. The apparatus (100) according to claim 10 or 11, wherein said control algorithm is based upon the following equation (or part of it): wherein is the right pseudo-inverse matrix of the geometric or analytical Jacobian (this choice depends upon the representation that is used to define the orientation) and are, respectively the error in speed and position expressed in the basic reference triad [X_B,Y_B,Z_B] The vectors are respectively the Cartesian position and speed error of the end member 11 of the manipulator 1 ; are the position and angular velocity of the same, respectively;

Ad₁ is the transposition of the added matrix that has been appropriately inserted in order to transform the position i and speed x error from the basic reference triad [X_B,Y_B,Z_B] to the reference triad of the end member of the robot [X_tool,Y_tool,Z_tool] and it is expressed as follows wherein is the rotation matrix that expresses the orientation of the reference triad of the end member, [X_tool,Y_tool,Z_tool] with respect to the basic one, [X_B,Y_B,Z_B], the symmetric skew matrix of the position of the end member, K_p and K_B are the matrices of the gains of the position control and they are 6 X 6 semi-defined positive diagonal matrices wherein are null so that the action of the control in position does not have effect along and around the tapping axis.

K_F is an identity matrix 6 x 6 and x_F is the control action which as a function of the quantity detected by said detecting means (21, 22, 24) allows said end member (11) to apply to the external element said effective quantity according to the tapping axis,

13. A control method of a robotic actuator (11 ) configured for the application of an effective quantity, selected in a group comprising torque, force, angular or linear speed, according to a working axis (A), which method provides to constrain the position and spatial orientation of said actuator (11) in the directions orthogonal to said working axis (A) so that they cannot be modified by the operator, which method provides to determine said effective quantity by means of a centralized control algorithm, for example an inverse dynamics algorithm, as a function of the force and/or operating torque exerted by an operator on a maneuvering portion (14) of said end member (11) and detected by detecting means (21, 22).

14. The method according to the preceding claim, wherein said effective quantity is calculated as proportional, directly or inversely, to said operating force or torque.

15. The method according to claim 13 or 14, wherein said control algorithm is based upon the following equation (or part of it): wherein is the right pseudo-inverse matrix of the geometric or analytical Jacobian (this choice depends upon the representation that is used to define the orientation) and are, respectively the error in speed and position expressed in the basic reference triad [X_B,Y_B,Z_B] The vectors are respectively the Cartesian position and speed error of the end member 11 of the manipulator 1 , On the contrary, are respectively the position and angular speed of the same.

Ad₁ is the transposition of the added matrix that has been suitably inserted in order to transform the position £ and speed x error from the basic reference triad [X_B,Y_B,Z_B] to the reference triad of the end member of the robot [X_tool,Y_tool,Z_tool] and it is expressed as follows wherein is the rotation matrix that expresses the orientation of the reference triad of the end member, with respect to the basic one, [X_B,Y_B,Z_B]. Whereas is the symmetric skew matrix of the position of the end member, P_{too l}K_p and k_D are the matrices of the gains of the position control and they are 6 x 6 semi-defined positive diagonal matrices wherein are null so that the action of the control in position does not have effect along and around the tapping axis.

K_F is an identity matrix 6 x 6 e X_F is the control action which as a function of the quantity detected by said detecting means (21, 22, 24) allows said end member (11) to apply to the external element said effective quantity according to the tapping axis.

16. The method according to any one of claims 13 to 15, which is applied for the control of an apparatus according to any one of claims 1 to 12.