EP4143654A1

EP4143654A1 - Robotic vehicle

Info

Publication number: EP4143654A1
Application number: EP21726187.4A
Authority: EP
Inventors: Luigi PALOPOLI; Daniele FONTANELLI; Fabiano ZENATTI; Stefano DIVAN; Marco Andreetto
Original assignee: Universita degli Studi di Trento
Current assignee: Universita degli Studi di Trento
Priority date: 2020-04-27
Filing date: 2021-04-19
Publication date: 2023-03-08
Also published as: WO2021220103A1; IT202000009040A1

Abstract

A robotic vehicle (10) comprising a pair of rear wheels (11, 12), at least one front wheel (13, 14), at least one steering motor (50) mounted with said at least one front wheel (13, 14), wherein said at least one front wheel (13, 14) is adapted to freely rotate about a vertical axis (Z_m) perpendicular to the ground, wherein said at least one steering motor (50) is adapted to impose a torque which steers said at least one front wheel (13,14) with which it is mounted making the robotic vehicle (10) steer by a steering angle (φ) with respect to an instantaneous direction of linear motion (X_m) of the robotic vehicle (10), wherein the robotic vehicle (10) comprises at least one processor (70) and at least one storage (80), wherein said at least one processor (70) controls said at least one steering motor (50) and calculates a spatial position and a spatial orientation of the robotic vehicle (10), wherein said at least one storage (80) contains at least one selection algorithm which implements a process for sharing a control authority for directing the robotic vehicle (10), wherein said at least one storage (80) contains a predefined preferred path (100) or the position of an obstacle to be overcome, wherein for each predefined time step said at least one selection algorithm selects depending on a metric a torque value which said at least one steering motor (50) imposes on said at least one front wheel (13, 14), wherein said torque value ranges between 0 and a predefined maximum torque value, wherein said metric is defined depending only on a distance of the robotic vehicle (10) from said predefined preferred path (100), or from an obstacle to be overcome, and on an angle of approach (δ), wherein the angle of approach (δ) of said metric is calculated by said at least one processor (70) depending on an approaching trajectory towards said predefined preferred path (100) or towards said obstacle to be overcome.

Description

ROBOTIC VEHICLE

The present invention relates to a robotic vehicle. In the state of the art, robotic vehicles such as robotic walkers for users with reduced walking ability are known.

Disadvantageously, robotic walkers help users with reduced walking ability to walk, but they do not make it possible to help users with reduced cognitive ability, e.g., patients with Alzheimer's disease or other types of cognitive impairments, who cannot remember for example where the room they are going to is or who cannot recognize obstacles in their path. Fully autonomous robotic walkers, i.e., those which control the entire path independently of the patient, are disadvantageously debilitating for patients who are still partially autonomous and whose residual abilities should be kept alive as much as possible.

On the other hand, for safety reasons it is not possible to leave the authority to control the movement direction of the robotic walker to the patient with reduced cognitive ability alone, otherwise the patient would risk stumbling or making the robotic walker collide with obstacles.

In the known state of the art, robotic walkers exist which are passive vehicles and which select, for each predefined time step, to entrust the authority to steer to the patient or to the robotic walker, disadvantageously subtracting from the patient the authority to control the direction of the robotic walker and leaving the perception to the intellectually healthy patient of being controlled by the robotic walker.

An example of a shared control system is described in the publication G. Wasson, P. Sheth, M. Alwan, K. Granata, A. Ledoux and Cunjun Huang, "User intent in a shared control framework for pedestrian mobility aids, " Proceedings 2003 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2003) (Cat. No.03CH37453), Las Vegas, NV, USA, 2003, pp. 2962-2967 vol.3, doi: 10.1109/IROS.2003.1249321.

The object of the present invention is to obtain a robotic vehicle which allows to share the control authority of the direction of motion between the robotic walker and the patient with reduced walking ability and/or reduced cognitive ability, overcoming the disadvantages of the known state of the art.

In accordance with the invention, such an object is reached by a robotic vehicle according to claim 1.

Another object of the present invention is to obtain a process for directing a robotic vehicle which allows to share the control authority of the direction of motion between the robotic vehicle and a user with impaired walking ability and/or reduced cognitive ability, overcoming the disadvantages of the known state of the art.

In accordance with the invention, such other object is achieved with a process for directing a robotic vehicle according to claim 7.

Other features are comprised in the dependent claims.

The features and advantages of the present invention will be more apparent from the following description, which is to be understood as exemplifying and not limiting, with reference to the appended schematic drawings, in which: figure 1 is a schematic top view of a particular robotic vehicle which is a robotic walker according to the present invention; figure 2 is a schematic side view of the robotic walker; figure 3 is a schematic view showing the robotic walker in three different positions and directions with respect to a predefined preferred path 100, where for each position an angle of approach and an approaching trajectory is shown; figure 4 shows a graph depicting an elastic gain value which essentially represents a torque to be applied to the steering motors of the robotic walker depending on an angular error tolerance with respect to the desired values to guide the robotic walker along a predefined preferred path 100 or away from an obstacle; and figure 5 is a schematic view of a robotic vehicle according to an embodiment of the invention, in which the coordinates l_x and l_y of the reference point O_m of the robotic vehicle are shown with respect to a Frenet reference frame positioned on the curvilinear abscissa s relative to a predefined preferred path.

With reference to the aforementioned figures and in particular figures 1 and 2 a particular robotic vehicle is shown which is a robotic walker 10 comprising four wheels 11-14, a pair of rear wheels 11, 12 and a pair of front wheels 13, 14.

The robotic walker 10 is a passive vehicle, i.e., a vehicle which is moved by the patient. The robotic walker 10 comprises at least one bar 30 which can be manoeuvred by a patient, in which the at least one manoeuvrable bar 30 is adapted to be pushed or pulled by the patient, such that the patient pushes or pulls the robotic walker 10. This avoids disadvantageous situations in which the robotic walker 10 may pull the patient at a higher velocity than the patient can sustain or may hinder the patient at a lower velocity than the patient would like to achieve.

The passive vehicle is moved by the patient and does not comprise any motor adapted to push or pull or move the robotic walker 10.

Preferably, the robotic walker 10 comprises a left-handed manoeuvrable bar and a right-handed manoeuvrable bar. Each manoeuvrable bar 30 comprises a knob 31 adapted to be gripped by the patient.

Preferably the two rear wheels 11, 12 are independent from each other.

The robotic walker comprises two electromechanical brakes 40 each of which is mounted with a wheel 11, 12 of the pair of rear wheels 11, 12.

Preferably, at least one of the two manoeuvrable bars 30 comprises a brake lever 41 adapted to allow the patient to apply at least one electromechanical brake 40 to brake at least one rear wheel 11, 12.

The two front wheels 13, 14 are not mounted on a fixed axis which joins them together and therefore each front wheel 13, 14 can freely rotate about a respective vertical axis Z_m which is perpendicular to the ground on which the front wheel 13, 14 rests.

Preferably, it is included that the electromechanical brake 40 can be a motor to brake the rear wheel 11, 12.

According to a preferred solution, it is further included that the brakes 40 can be independent so that they can function independently of each other to brake even only one of the two rear wheels 11, 12. In this preferred embodiment, it is included that each knob 31 can comprise a brake lever 41.

The robotic walker 10 comprises two steering motors 50 each of which is mounted on each of the two front wheels 13, 14, in which each steering motor 50 is adapted to steer the front wheel 13, 14 where it is mounted. The two steering motors 50 operate independently of each other so as to allow each of the front wheels 13, 14 to steer according to a different steering angle cp. Each of the two steering motors 50 imposes a torque on the respective front wheel 13, 14 with which it is mounted, so as to impose the steering of the front wheel 13, 14.

Advantageously, the two front wheels 13, 14 make the robotic walker 10 more stable and easier to manoeuvre with respect to a three-wheeled vehicle.

When a patient uses the robotic walker 10 and steers in one direction to rotate the robotic walker 10 about a centre of rotation 20, then a left front wheel 13 steers by a left steering angle cpi and a right front wheel 14 steers by a right steering angle cp_r, in which the two left steering angles cpi and right steering angles cp_r are different, as shown in figure 1. For the sole purpose of a simplification of the mathematical discussion concerning the steering of the robotic walker 10, a steering angle cp is considered which mathematically depends on the two left cpi and right cp_r steering angles and which is shown in figure 1 as the steering angle cp of a virtual front wheel 15 which is virtually arranged halfway between the left front wheel 13 and the right front wheel 14.

The steering angles are measured starting from a direction of linear motion along a direction X_m, which is an instantaneous direction of motion of the robotic walker 10.

The robotic walker 10 steers by the steering angle cp with respect to the direction of linear motion X_m which the robotic walker 10 had an instant before steering.

The robotic walker 10 comprises at least one processor 70 which controls electromechanical brakes 40 and steering motors 50 and which calculates a spatial position and orientation of the walker 10 with respect to the surrounding environment and/or with respect to the predefined preferred path 100 and/or with respect to obstacles and at least one storage 80 which contains commands, data, measurements, a selection algorithm to implement a process for sharing a control authority for directing the robotic walker 10.

Even more advantageously, the robotic walker 10 comprises a localization apparatus comprising at least one localization sensor and connected to the at least one processor 70 and at least one storage 80 of the robotic walker 10, for example the localization sensor can be a camera 60. For each predefined time range, the localization apparatus operated by the processor 70 by a localization process implemented in the storage 80 localizes the spatial position and orientation of the robotic walker 10 at least with respect to the surrounding environment, detects the position of obstacles to be overcome, measures a distance between the walker and the obstacles, calculates the predefined preferred path 100 of the robotic walker 10 according to predefined indications from third parties or from the patient, measures a distance between the robotic walker 10 and said predefined preferred path 100, measures the angle of approach d with respect to the predefined preferred path 100 and stores all these information in said at least one storage 80.

Preferably, the robotic walker 10 comprises an interface 90 which can be graphical or acoustic or a keyboard or other sensor device which is capable of collecting orders from the patient or a nurse or physician or family member indicating a direction or a location in the environment where the robotic walker 10 is to direct the motion of the patient, such as a room in the environment where the patient is present.

The robotic walker 10 advantageously shares the control authority with the patient by adjusting according to a control authority sharing process present in the storage 80 and driven by the processor 70.

The control authority sharing process of the direction of motion of the robotic walker allows a torque value to be selected for each time step to be applied to the steering motors 50 depending on a metric.

In other words, for each predefined time step, the selection algorithm selects depending, preferably only, on a metric a torque value which the at least one steering motor 50 imposes on the at least one front wheel 13, 14.

The torque value ranges between 0 and a predefined maximum torque value. The maximum torque value is predefined according to a calibration which verifies the mechanical features of the front wheels 13, 14 and the technical features of the steering motors 50

The metric is defined depending on a distance of the robotic walker 10 from a predefined preferred path 100 or obstacle to be overcome and depending on an angle of approach d.

In particular, the metric is defined depending only on a distance of the robotic vehicle 10 from the predefined preferred path 100, or an obstacle to be overcome, and an angle of approach d.

Such a feature allows selecting the torque to be imposed on the at least one front wheel 13, 14 regardless of the velocity of motion of the robotic vehicle 10.

Furthermore, a technical effect derived from the features of the robotic vehicle 10 according to the invention, in particular obtained by applying the selection algorithm contained in the storage 80, can be identified with the variable stiffness of the steering angle cp, i.e., with the modification of the torque to be applied to the steering, specifically to the at least one front wheel 13,14, of the robotic vehicle 10.

In addition, it should be noted that the robotic vehicle according to the invention allows only a torque of varying intensity to be imposed on the at least one front wheel 13, 14 by exploiting the natural physical interaction occurring at the at least one front wheel 13, 14 between the torque applied to the at least one front wheel 13, 14 and the torque generated by the user.

The robotic vehicle 10 does not need systems for estimating and/or measuring the torques impressed by the user on the vehicle itself, with consequent savings in sensors and computing power.

In contrast, the system described in Wasson G ET AL necessarily requires the presence of devices for estimating and measuring the torques impressed by the user on the system to merge or negotiate with the control torque of the system itself.

Preferably, the torque value imposed by the steering motor 50 at the at least one front wheel 13, 14 is proportional to the amount of a deviation from the metric.

As shown in figure 3, the angle of approach d is defined based on an approaching trajectory as considered for example in the PhD thesis entitled "Trajectory tracking for unicycle-type and two- steering-wheels mobile robots" by Alain Micaelli and Claude Samson published in INRIA, 1993, in the scientific paper "Adaptive, non singular path-following control of dynamic wheeled robots" by D. Soetanto, L. Lapierre and A. Pascoal published in "IEEE Conference on Decision and Control", volume 2, pages 1765-1770, IEEE, 2003 and in the scientific paper "Hybrid feedback path following for robotic walkers via bang-bang control actions" by Marco Andreetto, Stefano Divan, Daniele Fontanelli and Luigi Palopoli published in "Decision and Control CDC, 2016 IEEE 55th Conferences at pages 4855-4860, IEEE 2016.

For example, if the metric is the distance between the walker 10 and the predefined preferred path 100, then as the robotic walker 10 moves away from the predefined preferred path 100 and the angle of approach d results not directed in the direction of the predefined preferred path 100, the process selects an increasing torque value to be applied to the steering motor 50 to gradually remove the control authority from the patient and force the patient to follow the correct steering angle cp so as to return to the predefined preferred path 100 and according to a correct angle of approach d so that the patient and the robotic walker 10 can move safely.

For example, if the metric is the distance from the obstacle, then as the robotic walker 10 approaches the obstacle to be overcome and the angle of approach d is directed in the direction of the obstacle, then the process selects an increasingly larger torque value to apply to the steering motor 50 to gradually remove the control authority from the patient and force the patient to follow the correct steering angle cp so as to overcome the obstacle in a safe manner for both the patient and the robotic walker 10.

Preferably, the torque to apply to the steering motors 50 can be mathematically described by the following formula: where u_T is the torque which is applied to the steering motors 50, f is the calculated steering angle of the virtual wheel 15, f_ά is the desired steering value which should be applied to the virtual wheel 15 to optimize the metric, n_f is the calculated angular steering velocity of the virtual wheel 15, <p_d is the desired angular steering velocity which should be applied to optimize the metric, k_r and K_V are dimensionless values which represent an elastic gain and a damping gain, respectively, and can be varied depending on a tolerated angular error eg between the calculated values of f and n_f and the desired values p_d and <p_d . The tolerated error is between 0 sexagesimal degrees and a maximum value always expressed in sexagesimal degrees, where the maximum value is a predefined value of the tolerated angular error which is entered by a programmer within the selection algorithm.

The tolerated angular error eg represents an example of deviation from the metric mentioned earlier.

In particular, the tolerated angular error can be described mathematically by the following formula: eg = Q- 5(l_y) where Q is the difference between the orientation of the robotic vehicle 10 and the orientation of the predefined preferred path 100 and d is the angle of approach depending on l_y .

With reference to figure 5, l_x and l_y are the coordinates of the reference point O_m of the robotic vehicle 10 with respect to the tangent X_f and, respectively, normal Y_f vectors of a Frenet reference frame positioned on the curvilinear abscissa s relative to the predefined preferred path 100 (see, for example, the publication D Soetanto, L Lapierre, and A Pascoal, "Adaptive, non-singular path-following control of dynamic wheeled robots", IEEE Conf. on Decision and Control, volume 2, pages 1765-1770. IEEE, 2003). According to an aspect of the invention, the desired steering angle f_ά and the desired angular steering velocity f_ά are determined by applying, by means of the at least one processor 70, a control law which allows the robotic vehicle 10 to steer towards the predefined preferred path 100.

For example, the desired steering angle f_ά and the angular velocity thereof f_ά can be derived from the so- called backstepping control technique, ensuring the convergence of the vehicle 10 towards the predefined preferred path 100.

Specifically, the expression calculated using the backstepping control technique is as follows:

— F ^_ Fίί

4e₀ cos h = d cos(cp_d) ( c¾os(9) where, in addition to the above definitions,

• x is an auxiliary variable used in the backstepping process and only its first time derivative x is of interest to the controller,

• v is the longitudinal forward velocity of the vehicle 10 (measured),

• d is the pitch of the vehicle (depicted in figure 5),

• c(s) is the curvature of the predefined preferred path 100 (inverse of the radius of the osculating circle) calculated at the curvilinear abscissa s, and

• K_X > 0 and Kg> 0 are design constants, to be calibrated to modulate the accuracy of the metric tracking .

The gains k_r and K_V can be expressed by the following formulas: where Q represents the tolerable threshold orientation error value for eg and where k_p and k_v are two constants greater than zero which represent the maximum values for k_p and, respectively, k_v. Essentially when eg is equal to zero both gains k_r and K_v are zero, so the at least one steering motor 50 exerts no torque on the at least one front wheel 13,14 in addition to the torque applied by a user on the robotic vehicle 10 (totally passive robotic vehicle 10). On the other hand, when eg tends to Q, the two gains k_r and K_V tend to the maximum allowable value k_p and k_v and the maximum control action is exerted on the steering, in particular on the direction of the at least one front wheel 13,14, by the selection algorithm, thus making the steering be perceived as extremely stiff to a user moving the robotic vehicle

10.

The maximum values of the gains k_r and K_V depend on technical features of the front wheels 13, 14 and the respective steering motors 50 and can be found experimentally through a calibration of the system comprising the front wheels 13, 14 and the steering motors 50.

Figure 4 shows the trend of the elastic gain value K_p which essentially represents the value of the torque to apply to the steering motors 50. The graph shows the elastic gain k_r depending on the tolerated angular error eg with respect to the metric, where the selection algorithm requires that the maximum torque value is applied when the tolerated angular error is at most 45 sexagesimal degrees. This maximum tolerated angular error value is only an example and is chosen based on experimental calibration of the system of the front wheels 13, 14 and the steering motors 50. The trend of the damping gain K_v is similar to that shown for the elastic gain k_r and is not shown in the figures.

The formula used to describe the torque to apply to the steering motors 50 implies that the steering angle f and the angular steering velocity n_f of the virtual wheel 15 must be calculated by measuring the steering angle and angular steering velocity of both the left and right front wheels.

As can be seen from the formula, it is apparent that the torque to apply to the steering motors 50 does not depend on the velocity of motion of the robotic walker 10 imparted by the patient.

Advantageously, the robotic walker 10 according to the present invention allows to overcome the disadvantages of the state of the art solutions and to allow the patient to be able to lead the robotic walker 10 without having controlled movements, but experiencing an increased stiffness of the steering of the front wheels 13, 14 when the robotic walker 10 results outside of the metric, i.e., away from the predefined preferred path 100 or directed in the opposite direction. On the other hand, when the robotic walker 10 is directed towards the predefined preferred path 100 or is positioned along the predefined preferred path 100 within a tolerated angular error, then the patient will experience lower steering stiffness. Thereby, the patient is guided along the predefined preferred path 100 without being controlled by the robotic walker 10.

Alternatively, it is possible to include that the robotic walker 10 comprises three wheels, two rear wheels 11, 12 and a single front wheel 13 or 14 arranged in the virtual front wheel position 15 of figure 1 and that the front wheel steers the steering angle cp by means of a single steering motor 50.

Alternatively, the brakes 40 operate synchronously by braking both rear wheels 11, 12, or according to a further alternative solution a single motor 40 brakes both rear wheels 11, 12.

Alternatively, it is included that the electromechanical brake 40 can be a single mechanical brake comprising an electromechanical device which drives the mechanical brake which locks both rear wheels 11, 12.

Alternatively, it is included that the brakes 40 are not controlled by the processor 70 and are controlled only by the patient.

Alternatively, it is included that the robotic vehicle does not necessarily comprise brakes 40, but could comprise other safety systems for stopping the robotic vehicle.

Alternatively, it is included that the robotic vehicle 10 is braked by the two steering motors 50 by applying a maximum torque to the front wheels 13, 14 until the robotic vehicle 10 is stopped.

Alternatively, it is included that the robotic walker 10 does not comprise the localization apparatus but that this localization apparatus is external and detects the position of the robotic walker 10 in space or the position of obstacles and sends said position information to the storage 80 of the robotic walker 10 so that the position can be processed by the processor 70. The external localization apparatus can also measure the distance between the walker and obstacles, can calculate the predefined preferred path 100 of the robotic walker 10 according to predefined indications from third parties or from the patient.

Alternatively, it is possible to include that the robotic vehicle can also be an active vehicle and not only a passive vehicle, such as a wheelchair or a goods handling cart or other motorized vehicle driven by, for example, a user with cognitive impairments. In such a case, the robotic vehicle will be moved by a user- controlled motor, but the front wheels 13, 14 of the robotic vehicle will be controlled by the processor 70 of the robotic vehicle according to the present invention.

Alternatively, the process can be applied as a safety device of any vehicle if the driver loses the ability to manoeuvre the vehicle.

With respect to operation, a process is included for directing the robotic vehicle 10, in which a user moves the robotic walker 10 by means of the manoeuvrable bar 30, or alternatively the robotic vehicle 10 is motorized. The process comprises a first step of arranging the robotic vehicle 10 according to one of the above described embodiments, in which said at least one processor 70 calculates a spatial position and a spatial orientation of the robotic vehicle 10, a second step in which said at least one processor 70 calculates a distance of the robotic vehicle 10 from a predefined preferred path 100 or an obstacle to be overcome and calculates an angle of approach d according to an approaching trajectory towards said predefined preferred path 100 or towards said obstacle to be overcome, a third step in which said at least one processor applies the selection algorithm implementing the control authority sharing process to direct the robotic vehicle 10, in which the selection algorithm selects a torque value depending on the metric, a fourth step in which said at least one steering motor 50 imposes said selected torque on said at least one front wheel 13, 14.

It is further even more advantageously included that the first step includes that the localization apparatus operated by the processor 70 applies a localization process implemented in the storage 80 and localizes the spatial position and orientation of the robotic vehicle 10 at least with respect to the surrounding environment, detects obstacles, measures a distance between the robotic vehicle and the obstacles, calculates a predefined preferred path 100 of the robotic vehicle 10 according to predefined indications which are provided by third parties or by the patient, measures a distance between the robotic vehicle 10 and said predefined preferred path 100, measures the angle of approach d with respect to the predefined preferred path 100 and saves all this information in said at least one storage 80.

In the third step, the processor 70 implements the selection algorithm and selects the torque value depending on the metric, in which the torque value is as much greater than zero as the distance between the spatial position of the robotic walker 10 and the predefined preferred path 100 is minimum, and the spatial orientation of the robotic vehicle 10 is directed towards the angle of approach d to the predefined preferred path 100.

Alternatively, in the third step, the processor 70 implements the selection algorithm and selects the torque value depending on the metric, in which the torque value is as much greater than zero as the distance between the spatial position of the robotic vehicle 10 and the obstacle is minimum, and the spatial orientation of the robotic vehicle 10 is directed towards the angle of approach d to the obstacle.

Advantageously, the process for directing the robotic vehicle 10 overcomes the disadvantages of the state of the art.

The invention thus conceived is susceptible to many modifications and variants, all falling within the same inventive concept; furthermore, all details can be replaced by equivalent technical elements. In practice, the materials used, as well as the dimensions thereof, can be of any type according to the technical requirements .

Claims

1. A robotic vehicle (10) comprising a pair of rear wheels (11, 12), at least one front wheel (13,

14), at least one steering motor (50) mounted with said at least one front wheel (13, 14), wherein said at least one front wheel (13, 14) is adapted to freely rotate about a vertical axis (Z_m) perpendicular to the ground, wherein said at least one steering motor (50) is adapted to impose a torque which steers said at least one front wheel (13,14) with which it is mounted making the robotic vehicle (10) steer by a steering angle (cp) with respect to an instantaneous direction of linear motion (X_m) of the robotic vehicle (10), wherein the robotic vehicle (10) comprises at least one processor (70) and at least one storage (80), wherein said at least one processor (70) controls said at least one steering motor (50) and calculates a spatial position and a spatial orientation of the robotic vehicle (10), wherein said at least one storage (80) contains at least one selection algorithm which implements a process for sharing a control authority for directing the robotic vehicle (10), wherein said at least one storage (80) contains a predefined preferred path (100) or the position of an obstacle to be overcome, wherein for each predefined time step said at least one selection algorithm selects depending on a metric a torque value which said at least one steering motor (50) imposes on said at least one front wheel (13, 14), wherein said torque value ranges between 0 and a predefined maximum torque value, wherein said metric is defined depending only on a distance of the robotic vehicle (10) from said predefined preferred path (100), or from an obstacle to be overcome, and on an angle of approach (d), wherein the angle of approach (d) of said metric is calculated by said at least one processor (70) depending on an approaching trajectory towards said predefined preferred path (100) or towards said obstacle to be overcome.

2. The robotic vehicle (10) according to claim 1, wherein said torque value imposed by said steering motor (50) on said at least one front wheel (13, 14) is proportional to the amount of a deviation from said metric.

3. The robotic vehicle (10) according to claim 1 or 2, characterized in that it comprises two front wheels (13, 14), two steering motors (50), each of said two steering motors (50) is mounted with each of said two front wheels (13, 14).

4. The robotic vehicle (10) according to claim 3, characterized in that the two steering motors (50) are independent from each other.

5. The robotic vehicle (10) according to any one of claims 1-4, characterized in that it comprises a localization apparatus comprising at least one localization sensor and connected to the at least one processor (70) and to the at least one storage (80) of the robotic vehicle (10), wherein for each predefined time step the localization apparatus operated by the processor (70) by a localization process implemented in the storage (80) localizes the spatial position and orientation of the robotic vehicle (10) at least with respect to the surrounding environment, identifies the position of the obstacles to be overcome, measures a distance between the vehicle and the obstacles, calculates a predefined preferred path (100) of the robotic vehicle (10) depending on predefined indications provided by third parties or by the patient, measures a distance between the robotic vehicle (10) and said predefined preferred path (100), measures the angle of approach (d) with respect to the predefined preferred path (100) and stores all this information in said at least one storage (80).

6. The robotic vehicle (10) according to any one of claims 1-5, characterized in that it is a passive vehicle comprising at least one bar (30) which can be manoeuvred by a user, wherein said manoeuvrable bar (30) is adapted to be pushed or pulled by the patient for moving the robotic vehicle (10).

7. The robotic vehicle (10) according to any one of claims 1-6, characterized in that it is a robotic walker (10).

8. A process for directing a robotic vehicle (10), wherein the process comprises a first step of arranging a robotic vehicle according to any one of claims 1 to 7, wherein said at least one processor (70) calculates a spatial position and a spatial orientation of the robotic vehicle (10), a second step wherein said at least one processor (70) calculates the distance of the robotic vehicle (10) from a predefined preferred path (100) or from an obstacle to be overcome and calculates the angle of approach (d) depending on an approaching trajectory towards said predefined preferred path (100) or towards said obstacle to be overcome, a third step wherein said at least one processor applies the selection algorithm which implements the process of sharing the control authority for directing the robotic vehicle (10), wherein the selection algorithm selects a torque value depending on the metric, a fourth step wherein said at least one steering motor (50) imposes said selected torque to said at least one front wheel (13, 14).

9. The process according to claim 8, wherein the robotic vehicle (10) is arranged according to claim 5 and the first step includes that the localization apparatus operated by the processor (70) applies a localization process implemented in the storage (80) and localizes the spatial position and orientation of the robotic vehicle (10) at least with respect to the surrounding environment, identifies the position of the obstacles to be overcome, measures a distance between the robotic vehicle (10) and the obstacles, calculates a predefined preferred path (100) of the robotic vehicle (10) depending on predefined indications provided by third parties or by the user, measures a distance between the robotic vehicle (10) and said predefined preferred path (100), measures the angle of approach (d) with respect to the predefined preferred path (100) and stores all this information in said at least one storage (80).

10. The process according to any one of claims 8 or 9, characterized in that in the third step the processor (70) implements the selection algorithm and selects the torque value depending on the metric, wherein the torque value is as much greater than zero as the distance between the spatial position of the robotic vehicle (10) and the predefined preferred path (100) is minimum, and the spatial orientation of the robotic vehicle (10) is directed towards the angle of approach (d) to the predefined preferred path (100).

11. The process according to any one of claims 8 or 9, characterized in that in the third step the processor (70) implements the selection algorithm and selects the torque value depending on the metric, wherein the torque value is as much greater than zero as the distance between the spatial position of the robotic vehicle (10) and the obstacle is minimum, and the spatial orientation of the robotic vehicle (10) is directed towards the angle of approach (d) of the obstacle.