WO2022124938A1

WO2022124938A1 - Robot motion planning method and mobile robot

Info

Publication number: WO2022124938A1
Application number: PCT/RU2021/050336
Authority: WO
Inventors: Михаил Вячеславович ЕГОРЧЕВ; Андрей Андреевич ЗАВОДСКОВ
Original assignee: Общество с ограниченной ответственностью "РобоСиВи"
Priority date: 2020-12-07
Filing date: 2021-10-10
Publication date: 2022-06-16
Also published as: RU2749202C1

Abstract

A method for planning the motion of a robot in an area of permissible motion on a surface includes discretizing the state space of a robot and determining a finite set of elementary trajectories connecting pairs of states of the robot. For each elementary trajectory, the following is determined: a safe zone based on the visibility of safety sensors, a basic outline combining the safe zone with the dimensions of the robot, and an elementary trajectory outline combining the basic outlines of each intermediate state of a trajectory. A directed weighted multigraph is then constructed without loops, where the vertices correspond to discrete states of the robot, the arcs correspond to elementary trajectories lying in the area of permissible motion, and the weight of an arc corresponds to the time of travel along a trajectory, and the motion of the robot is planned by searching the multigraph for the path having the lowest sum of weights. This increases fluidity of motion and reduces the travel time of the robot along its course.

Description

ROBOT PLANNING METHOD AND MOBILE ROBOT

FIELD OF TECHNOLOGY TO WHICH THE INVENTION RELATES

The present invention relates to robotics, in particular, to a method for planning the movement of a robot in the area of permissible movement on a plane, as well as to a mobile robot with the possibility of implementing this method.

BACKGROUND OF THE INVENTION

Planning the movement of a robot is one of the main tasks of mobile robotics. The goal of planning is to select trajectories that simultaneously satisfy the criteria of optimality (for example, to achieve the minimum distance and / or time of movement) and safety, i.e. to avoid collisions of the robot with obstacles. Many solutions have been created in this area, including various solutions for planning the trajectories of robots in a certain area of admissible movement on a plane, for example, in a warehouse. Within the limits of the area of permissible movement, the appearance and disappearance of static obstacles is a fairly rare event. In turn, the space outside this region contains with a high probability static obstacles that are constant for the environment under consideration.

It is assumed that the robot has one driving wheel or its model can be reduced to a model with one driving wheel, the slipping of which can be neglected. It is also assumed that the robot is equipped with safety sensors that check for obstacles in real time in a certain area of the field of view, determined by the angle of rotation and the speed of the drive wheel, hereinafter referred to as the "safety zone". When an obstacle is detected in the safety zone, the sensor limits the speed of the robot accordingly. Due to the fact that the number of different safety zones is finite, each of them corresponds to a range of speeds and wheel angles, and the speed limits imposed when an obstacle is detected in it correspond to the worst case. The robot is additionally equipped with sensors for preventive detection of obstacles, also operating in real time, the field of view of which exceeds the field of view of the robot's safety sensors.

Classical methods for planning the movement of a mobile robot are based on constructing optimal trajectories in the area of acceptable movement, for example, with using algorithms such as A * and PRM (probalistic roadmap - probabilistic road maps), taking into account the dimensions of the robot to safely avoid obstacles.

However, the application of this approach leads to the fact that during the planned movement, the robot may be near the border of the area of admissible movement, having a high speed. At the same time, the zones checked by the safety sensors may go beyond the border of the area of admissible movement, where the robot's safety sensors are likely to detect obstacles and limit the speed of the robot.

A situation similar to the one described above occurs when driving around static obstacles located near the trajectory built based on the robot’s own dimensions in the allowable movement area: the robot, approaching the obstacle at high speed, abruptly dropped it due to the operation of safety sensors.

Such solutions, based on the classical method of taking into account the dimensions of the robot when planning the trajectory, have several disadvantages: firstly, the lack of consideration for the decrease in speed at the stage of choosing the trajectory of motion leads to finding solutions that are not optimal in time; secondly, the movement of the robot ceases to be smooth due to frequent reset and acceleration. The latter drawback can be particularly disadvantageous for warehouse robots carrying loads, often loose ones, that can shift and even fall off the loading platform when the robot brakes and accelerates.

A known method for planning the movement of autonomous mobile robots based on safe space (Safe Space), described in the article Park J.-H., Huh U.-Y. Path Planning for Autonomous Mobile Robot Based on Safe Space // Journal of Electrical Engineering and Technology. 2016. Vol. 11, no. 5, pp. 1441-1448. URL: https://doi.org/10.5370/JEET.2016.l L5.1441 (accessed 11/16/2020). The known method is based on the definition of the state space of the robot, which includes all the kinematic states of the robot, in which planning is carried out. For this purpose, a safe space is defined in the state space of the robot, which is defined by a set of states of the robot, in which the distance from the robot to all obstacles surrounding it is safe, i.e. includes the minimum distance required to stop the robot before it collides with an obstacle, taking into account the limited field of view of the robot's safety sensors. The distance to stop the robot is calculated taking into account the distance traveled by the robot during the time it takes to recognize an obstacle and start braking, as well as the braking distance from the start of braking to the complete stop of the robot. In addition, the method takes into account the risk of hidden obstacles in areas outside the visibility of the robot's safety sensors, in particular, in areas hidden from the safety sensors by stationary obstacles. For this purpose, a risk distance is calculated corresponding to the minimum safe distance from the robot to the risk point at which an obstacle may appear. Thus, areas of risk are set around stationary obstacles, which impose restrictions on the planning of the robot's movement: it is required either to limit the speed of the robot's movement, or to build a safe trajectory of the robot's movement around the risk area. Further, in a known method, a globally optimal trajectory of the robot is selected using the A* algorithm with a limitation on the speed of the robot according to the cost of risk.

The introduction of a safe space in the described method can significantly increase the smoothness of the robot's movement, compared with the classical approach, based solely on taking into account the dimensions of the robot. However, the introduction of risk areas around all stationary obstacles leads to non-optimal results of planning the movement of the robot, especially in the conditions of a known area of admissible movement, where the appearance and disappearance of obstacles is a rare event, thus, the known method can lead to excessive restrictions on the speed of the robot. However, the described method is the closest solution to the claimed invention.

The technical problem to be solved by the invention is to plan the movement of a nonholonomic robot for some initial and final configurations of the robot within a given area of admissible movement on the plane, with the choice of globally optimal trajectories, bypassing static obstacles, and also taking into account potential speed limits, input by security sensors.

DISCLOSURE OF THE INVENTION

To solve this problem, a method for planning the movement of a robot in the area of permissible movement on a plane is proposed, which includes the following stages: discretization of the state space of the robot by introducing a grid with a fixed step into the area of permissible movement and choosing a finite set of rotation angles, specifying a finite set of elementary trajectories in local coordinate system of the robot, determined by the angle of rotation of the driving wheel, its speed and duration of movement, and connecting pairs of different discrete states of the robot in the specified space, for each elementary trajectory: a safety zone is set based on the visibility zone of the robot's safety sensors at a given angle of rotation and speed of the driving wheel, a basic contour is set by combining safety zones with the dimensions of the robot, and set the contour of the elementary trajectory by combining the basic contours placed in each intermediate state of the elementary trajectory, constructing an oriented weighted multigraph without loops, the vertices of which correspond to the discrete states of the robot, the arcs correspond to the elementary trajectories, and the weight of the arc corresponds to the time the robot moves along the elementary trajectory, and in the graph includes only arcs, the contours of the elementary trajectories of which lie completely in the given area of admissible motion, and only those vertices from which at least one arc proceeds, and determine the sequence of elementary paths thorium for moving the robot to the target state by searching the constructed multigraph for the path with the smallest sum of weights.

Thanks to the described approach to the introduction and consideration of safety zones, the claimed method provides a smoother and time-optimized movement of the robot. In particular, when using the invention, the following effects are achieved: the robot maintains a large distance from the border of the area of permissible movement when it develops a high speed; when moving along a narrow section, the robot deliberately reduces speed; turns at "crossroads", at the boundaries of the zone of permissible movement, the robot performs more smoothly and at high speed.

According to an embodiment of the invention, the claimed method further comprises a step in which the basic contour of each elementary trajectory is expanded by the amount of the error in the movement of the robot along the trajectory. This makes it possible to take into account the error of the robot's movement along the selected trajectory, if the error is present and does not exceed a certain threshold value. For example, in this way it is possible to take into account the error introduced by the slippage of the driving wheel of the robot, or the errors introduced by the positioning system. According to the following embodiment of the invention, the claimed method further comprises the steps of: moving the robot according to the selected sequence of elementary trajectories and their speeds, checking the area surrounding the robot for the presence of static obstacles in real time using preventive obstacle detection sensors, checking the contours of the selected sequence of elementary trajectories for the presence of intersection with the detected static obstacles, and in the presence of the specified intersection, a new sequence of elementary trajectories is determined for the movement of the robot to the target state by searching in the constructed multigraph for a new path with the smallest sum of weights, while all arcs, contours of elementary trajectories are excluded from the search which intersect with the detected static obstacles, and the search of multiple arcs is carried out in ascending order of weight, until the arc is detected, the contour of the elementary trajectory which does not intersect with the detected static obstacles.

The invention in the embodiment described above makes it possible to rebuild the trajectory of the robot in case of detection of obstacles by sensors of preventive obstacle detection. Rebuilding is performed on the basis of the same multigraph built earlier, which significantly reduces computational costs and increases the speed of rebuilding.

According to an additional embodiment of the invention, the field of view of the robot's preventive obstacle detection sensors exceeds the field of view of the robot's safety sensors, so that the definition of a new sequence of elementary trajectories for the robot to move to the target state due to the detection of a static obstacle by the preventive obstacle detection sensors is carried out without limiting the speed of the robot.

The difference between the visibility areas of the sensors for preventive detection of obstacles and the safety sensors can be selected based on the maximum speed of the robot, so that in all cases the trajectory is rebuilt before the safety sensors are triggered and, accordingly, the speed of the robot is limited. Thus, the invention makes it possible to rebuild the trajectory of the robot in real time and without reducing the speed of the robot. The solution to the problem was also achieved with the help of a mobile robot containing: a chassis with a drive wheel to ensure the movement of the robot with a given speed range, safety sensors to detect obstacles in the field of view of the safety sensors surrounding the robot, and a control system communicatively connected with the safety sensors, made with the ability to scheduling the movement of the robot in the area of allowed movement on the plane according to the method described above, and controlling the chassis to carry out the movement of the robot.

According to the invention, a smoother and time-optimized movement of the claimed mobile robot is provided.

In a further embodiment, the mobile robot comprises proactive obstacle detection sensors communicatively coupled to the control system to detect obstacles in a field of view of the proactive obstacle detection sensors surrounding the robot that exceeds the field of view of the safety sensors of the robot. At the same time, the control system of the robot can be configured to plan the movement of the robot in the area of allowable movement on the plane according to the embodiments of the claimed method, which provides for checking the area surrounding the robot for the presence of static obstacles using preventive obstacle detection sensors. Accordingly, the invention makes it possible to rebuild the trajectory of the mobile robot in real time and without reducing the speed of movement.

Finally, the invention provides that in all embodiments, the mobile robot can be used to solve various kinds of problems, in particular, to solve warehouse logistics problems, and can be a robotic stacker, reach truck, tractor, warehouse cart, etc. and.

Each of the embodiments of the invention described above makes it possible to achieve a technical result, which consists in optimizing the motion planning of a nonholonomic robot for some initial and final robot configurations within a given area of admissible movement on a plane, including by increasing the smoothness of movement and reducing the time of movement of the robot along route.

BRIEF DESCRIPTION OF THE DRAWINGS

The essence of the invention is explained below on the example of some variants of its implementation with reference to the drawings containing the following illustrations. 1 shows a robotic electric stacker in which the present invention may be implemented.

2 is a schematic representation of a stacker robot in which the present invention is implemented, global and local coordinate systems.

Fig.3 - set of rotation angles of the robot stacker, given according to the invention.

Fig.4 is an example of an elementary trajectory traversed by the reference point of the local coordinate system at a fixed angle of rotation of the drive wheel of the stacker robot.

5 is an exemplary selection of a plurality of elementary trajectories of a stacking robot according to the invention.

Fig.b is an example of a set of contours of elementary trajectories that connect the same initial and final states, but corresponding to different speeds of the stacker robot.

Fig.7 - another example of a set of contours of elementary trajectories, linking the same initial and final states, but corresponding to different speeds of the stacker robot.

Fig.8 - the result of planning the movement of the stacker robot in the area of admissible movement in the classical way, based on the dimensions of the robot, without taking into account the contours of the elementary trajectories.

Fig. 9 is a result of motion planning of the stacker robot in the allowable motion area of Fig. 8 according to an embodiment of the invention.

Fig.10 - the result of planning the movement of the stacker robot in the area of admissible movement in the classical way, based on the dimensions of the robot, without taking into account the contours of elementary trajectories, in case of detection of static obstacles.

Fig. 11 is a result of motion planning of the stacking robot in the allowable motion area of Fig. 10 according to an embodiment of the invention.

IMPLEMENTATION OF THE INVENTION

The invention is aimed at optimizing the planning of the movement of a robot equipped with safety sensors and sensors for preventive obstacle detection in the area of permissible movement on a plane. Within the area of acceptable movement, the appearance and disappearance of static obstacles is considered a rare event. The concept of the area of permissible movement is applicable, for example, to a warehouse with a known geometry, as well as to many other spaces, subject to measurement and control. The space outside the area of admissible movement contains with a high probability static obstacles that are constant for the considered environment.

For clarity, it is assumed that the robot has one drive wheel, the slip of which can be neglected. An example of such a robot is the robotic electric stacker 1 shown in FIG. 1, in which the present invention has been implemented. Hereinafter, the terms "robot", "mobile robot", "stacker robot", "stacker" are used interchangeably and refer to the robot in which the present invention is implemented, unless otherwise expressly indicated. At the same time, it should be understood that the invention is not limited in this regard and can be successfully implemented on the basis of any mobile robot, the kinematic model of which is reducible to the kinematic model of a robot with one driving wheel, in particular, in various robotic warehouse vehicles, such as tractor, storage trolley, reach truck, etc. .

The safety sensors 2 of the robot check in real time the presence of obstacles in a certain area of visibility, determined by the angle of rotation and the speed of the drive wheel 3, called the safety zone. When an obstacle is detected in the safety zone, the safety sensor limits the speed of the robot accordingly. Proactive obstacle detection sensors 4 robots also work in real time and have areas of view that exceed those of safety sensors. In this case, as shown in Fig.1, the sensor part of the sensor system of preventive obstacle detection 4 can be at least partially combined with the sensor part of the safety sensor system 2, provided that each of the safety sensor systems and preventive detection of obstacles has its own data processing subsystems and executive mechanisms.

On the considered plane, two Cartesian right-handed coordinate systems are introduced: a fixed, global coordinate system and associated with the drive wheel 3 of the robot, the local coordinate system, as shown in Fig.2 using the example of the stacker robot 1. Thus, the position of the robot on the plane, hereinafter referred to as the state of the robot can be represented by three values: displacement and rotation of the reference point of the local coordinate system relative to the global one. The set of all possible states of the robot in the area of permissible movement forms the state space of the robot. As mentioned above, the robot has one drive wheel or can be represented by a kinematic model with one drive wheel, the range of possible rotation angles is known, as well as the range of possible speeds. It is assumed that the slippage of the driving wheel of the robot can be neglected, and a kinematic model of motion is adopted. These assumptions are relevant and justified for a wide range of tasks, in particular, for planning the movement of robots in enclosed spaces, such as warehouses. Moreover, practice shows that the above assumptions can also be made for planning the movement of mobile robots in open spaces. In these cases, errors due to the possible slippage of the driving wheel of the robot can be neglected, or they can be compensated using the robot's control system that tracks its location and speed. At the beginning of the robot motion planning method of the present invention, the state space of the robot is discretized: a grid with a fixed step is introduced in the area of permissible motion, and a finite set of robot rotation angles is selected. An example of a plurality of rotation angles 5 selected for a stacker robot 1 is illustrated in FIG.

Thus, the discrete state of the robot is represented by the indices of the nearest grid vertex in the area of permissible movement and the index of the nearest rotation angle. Corresponding trajectories traversed by the point of reference of the robot's local coordinate system consist of line segments, circular arcs and turning points in place. Each separate section of such a trajectory, further called an elementary trajectory, is determined by three parameters: the angle of rotation of the drive wheel, its speed, and the duration of movement. An example of an elementary trajectory 6, which is a path traversed by the reference point of the local coordinate system for a certain period of time at a fixed angle of rotation 7 and the speed V of the drive wheel 3 of the stacker robot 1 is shown in Fig.4.

With this in mind, in the method according to the invention, a finite set of elementary trajectories are set with sufficient accuracy connecting pairs of different discrete states of the robot in the state space of the robot. In Fig.5 you can see an example of a selection from the set of elementary trajectories 8 of the robot stacker 1.

The angle of rotation of the robot wheel and its speed completely determine the field of view of the safety sensors when the robot moves along an elementary trajectory. With this in mind, in the method according to the invention, for each elementary trajectory, a safety zone is set based on the field of view of the robot's safety sensors. given the angle of rotation and the speed of the drive wheel. Further, for each elementary trajectory, a basic contour is set, which is a combination of safety zones with the dimensions of the robot. Finally, for each elementary trajectory, the contour of the elementary trajectory is set, which is the union of the basic contours placed in each intermediate state of the elementary trajectory. Several examples of the contours of the elementary trajectories of the stacker robot 1 are shown in Fig.b and 7. In each of the drawings in Fig.b and 7, the contours 9, 10, 11, 12 and 13 of elementary trajectories connecting the same initial and final states are illustrated. , but corresponding to different speeds of movement of the robot stacker 1; and also shows the path 14, which in these cases passes the reference point of the local coordinate system of the robot. On each of the figures, contour 9, marked with triangles, corresponds to the speed of the stacker robot in 0.3 m/s; contour 10, marked with square signs, corresponds to a speed of 0.7 m/s; contour 11, marked with a straight cross, corresponds to a speed of 1.0 m/s; contour 12, marked with diamond signs, corresponds to a speed of 1.38 m/s; and contour 13, marked with signs of an inclined cross, corresponds to a speed of 1.67 m/s.

Since the robot can move along this trajectory with some error, in an additional embodiment of the invention, the basic contour of each elementary trajectory is expanded by the amount of such an error. In this embodiment, the invention can be used, among other things, for planning the movement of the robot in open spaces where slippage of the driving wheel of the robot is possible.

The given area of allowed movement of the robot in the global coordinate system can be represented, for example, by a set of flat polygons. This area is fixed, and to take into account the possibility of short-term occurrence of obstacles in it, data from sensors of preventive obstacle detection is used.

According to the claimed method, an oriented weighted multigraph without loops is built once, hereinafter referred to simply as a multigraph for brevity, the vertices of which correspond to the discrete states of the robot, and the arcs correspond to elementary trajectories. The multigraph includes only those arcs whose elementary trajectory contours lie completely in the region of admissible motion, as well as only those vertices from which at least one arc originates. As the weights of the arcs, the time of movement along elementary trajectories is used. Thus, the path in the multigraph with the smallest sum of weights on the arcs will allow reaching the target state in least time. Accordingly, the robot motion planning method of the present invention comprises determining a sequence of elementary trajectories for the robot to move to a target state by searching the constructed multigraph for the path with the smallest sum of weights.

Pathfinding in a multigraph can be performed using various well-known algorithms such as A*, D*, RTAA*, LSS-LRTA*, etc. The invention is not limited to an algorithm for finding a path in a multigraph, and any appropriate algorithm, including including algorithms that may be developed in the future, can be successfully applied in all embodiments of the present invention.

Multiple arcs in the multigraph correspond to elementary trajectories with the same angle of rotation of the drive wheel and the same length, but different speeds and different travel times, i.e. different weight. It is obvious that the arc with the highest speed (least weight) will be the most preferable among all multiple arcs, however, since it also corresponds to the largest contour, it can be excluded from consideration in the process of rebuilding the path during the movement of the robot, if it is found to intersect with such obstacle outline.

To do this, in the embodiment of the invention, the method additionally comprises the steps of: moving the robot according to the selected sequence of elementary trajectories and their speeds, and in real time checking the area surrounding the robot for the presence of static obstacles using preventive obstacle detection sensors. The contours of the selected sequence of elementary trajectories are checked for the presence of intersection with the detected static obstacles, and in the presence of the specified intersection, a new sequence of elementary trajectories is determined for the robot to move to the target state by searching in the constructed multigraph for a new path with the smallest sum of weights, while all arcs are excluded from the search, the contours of the elementary trajectories of which intersect with the detected static obstacles, and the enumeration of multiple arcs is carried out in order of increasing weight, until an arc is found, the contour of the elementary trajectory of which does not intersect with the detected static obstacles. The last rule for enumeration of multiple arcs increases the efficiency of the claimed method, since it increases the speed of finding a new path in a multigraph without worsening the search results. At the same time, according to a preferred embodiment of the invention, the field of view of the robot's preventive obstacle detection sensors, which exceeds the field of view of the robot's safety sensors, based on the indications of which the robot's speed can be limited, is selected in such a way that the definition of a new sequence of elementary trajectories for the robot to move to the target state due to the detection of a static obstacle by sensors, preventive obstacle detection is carried out without limiting the speed of the robot. In other words, the difference in the field of view of the preventive obstacle detection sensors and the safety sensors of the robot is selected so that after the preventive obstacle detection sensors are triggered, the robot has time to rebuild and change the movement trajectory before the safety sensors are triggered, even in the case of movement at maximum speed.

As mentioned above, the invention was implemented in the hardware and software system of the robotic electric stacker 1 shown in Fig.1. The specified stacker 1 is a mobile robot, which is a separate object of the present invention. The implementation of the invention based on the stacker robot 1 is described below in detail to provide an understanding of the essence of the invention. However, it should be understood that all technical details, meanings and values presented are merely examples of one or more embodiments of the invention, and in no way limit the scope of the legal protection provided by the claims.

As shown in Fig.1, the stacker robot 1 is equipped with two laser scanners located in its front part on the sides and combining the functions of the sensor part of the safety sensors 2 and sensors of preventive obstacle detection 4. Combining the sensor part of safety sensors and preventive obstacle detection is preferable to increase the compactness and reduce the cost of the robot equipment, but is not the only possible solution. With equal success, safety sensors and sensors for preventive obstacle detection can be implemented as separate devices or modules.

The stacker robot 1 has one driving wheel 3, its rotation angle (p lies in the range of [-90, 90] degrees, and the linear speed V lies in the range of [-0.3, 1.67] m/s, for example, as shown in Fig. 4. The wheelbase L is 1.386 m. Based on the parameters of the problem, at the stage of discretization of the state space on the plane, a grid was introduced with a step of 0.1 m along the axes OX and OY of the global coordinate system shown in Fig.2, and also fixed set 5 of 24 rotation angles of the driving wheel 3 of the robot, shown in Fig.Z. Next, a set of 918 elementary trajectories was recorded, a sample of 8 of which is shown in Fig.5 for example. The invention is not limited in respect of all the specified parameters, the grid spacing, the number of discrete rotation angles and elementary trajectories of the robot can be specified differently, depending on the conditions of the problem being solved, the required accuracy, available computing power, etc.

Based on the above parameters, the method of planning the movement of the stacker 1 according to the invention was applied. Figures b and 7 show examples of the contours of elementary trajectories calculated for the stacker robot 1: in each case, a set of contours of the elementary trajectory 9, 10, 11, 12 and 13 is shown, linking the same initial and final states, but corresponding to different speeds of the robot, from 0.3 m/s to 1.67 m/s, as mentioned above.

Based on these parameters, the movement of the stacker 1 was planned in various areas of permissible movement by the method according to the embodiments of the invention, as well as by the classical method, without taking into account the safety zones and contours of the elementary trajectories described above, which allows comparing the planning results.

Figure 8 shows an example of planning the movement of the stacker robot 1 from point A to point B in a given area of permissible movement with boundaries 15 in the classical way, based on the dimensions of the stacker robot 1. Figure 9 shows the result of applying an embodiment of the present invention for planning movement of the robot stacker 1 under conditions similar to those shown in Fig.8. Another example of planning the movement of the robot stacker 1 in the classical way is shown in Fig.10, in this case, static obstacles 16 are detected during the movement of the robot in the area of permissible movement. FIG. 11 shows the result of applying an embodiment of the present invention to planning the motion of a stacking robot 1 under conditions similar to those shown in FIG. 10. FIG.

Comparison of the robot motion trajectories in Figs. 8 and 9, 10 and 11 clearly shows that by optimizing the planning of the robot motion in the embodiments of the present invention, the following beneficial effects are achieved: the robot maintains a greater distance from the boundary of the allowable motion area when it develops a high speed; when moving along a narrow section, the robot deliberately reduces speed; turns at "crossroads", at the boundaries of the zone of permissible movement, the robot performs more smoothly and at high speed. The embodiments described above are exemplary only, the specific parameters, values and quantities used therein should also be construed as non-limiting examples. After reading the present description, specialists in the art may suggest various changes and adjustments to the described embodiments without changing the essence of the claimed invention. All such changes and modifications fall within the scope of the legal protection of the claims.

Claims

CLAIM

1. A method for planning the movement of a robot in the area of allowable movement on a plane, comprising the steps at which: the state space of the robot is discretized by introducing a grid with a fixed step into the area of allowable movement and choosing a finite set of rotation angles, specifying a finite set of elementary trajectories in the local coordinate system of the robot, determined by the angle of rotation of the driving wheel, its speed and duration of movement, and connecting pairs of different discrete states of the robot in the specified space, for each elementary trajectory: a safety zone is set based on the field of view of the robot's safety sensors at a given angle of rotation and speed of the driving wheel, define the basic contour by combining the safety zones with the dimensions of the robot, and define the contour of the elementary trajectory by combining the basic contours placed in each intermediate state of the elementary trajectory, build an oriented weighted multigraph without loops, whose vertices correspond to the discrete states of the robot, the arcs correspond to elementary trajectories, and the weight of the arc corresponds to the time of the robot’s movement along the elementary trajectory, and the graph includes only arcs whose elementary trajectory contours lie completely in the given region of admissible motion, and only those vertices from which at least one arc emanates and determine the sequence of elementary trajectories for the robot to move to the target state by searching the constructed multigraph for the path with the smallest sum of weights.

2. The method according to claim 1, further comprising the step of: expanding the basic contour of each elementary trajectory by the amount of the error in the movement of the robot along the trajectory.

3. The method of claim 1 or 2, further comprising the steps of: move the robot according to the selected sequence of elementary trajectories and their speeds, in real time check the area surrounding the robot for the presence of static obstacles using preventive obstacle detection sensors, check the contours of the selected sequence of elementary trajectories for the presence of intersection with the detected static obstacles, and in the presence of the specified intersections determine a new sequence of elementary trajectories for the robot to move to the target state by searching in the constructed multigraph for a new path with the smallest sum of weights, while all arcs are excluded from the search, the contours of elementary trajectories of which intersect with the detected static obstacles, and the multiple arcs are searched in ascending order weight, until the detection of an arc, the contour of the elementary trajectory of which does not intersect with the detected static obstacles.

4. The method according to claim 3, in which the visibility area of the robot's preventive obstacle detection sensors exceeds the visibility area of the robot's safety sensors, so that the determination of a new sequence of elementary trajectories for the robot to move to the target state due to the detection of a static obstacle by the preventive obstacle detection sensors is carried out without speed limitation robot movements.

5. A mobile robot comprising: a chassis with a driving wheel to ensure the movement of the robot with a given speed range, safety sensors to detect obstacles in the field of view of the safety sensors surrounding the robot, and a control system communicatively connected with the safety sensors, made with the possibility of planning the movement of the robot in the area allowable movement on the plane according to the method according to paragraph 1 or 2, and control of the chassis to carry out the movement of the robot.

6. The mobile robot according to claim 5, additionally containing preventive obstacle detection sensors communicatively connected to the control system for detecting obstacles in the visibility area of the preventive obstacle detection sensors surrounding the robot, which exceeds the visibility area of the robot's safety sensors.

7. Mobile robot according to claim 6, in which the control system is configured to plan the movement of the robot in the area of admissible movement on the plane according to the method according to paragraph 3 or 4.

8. Mobile robot according to any one of pi.