WO2020174052A1 - Traversing a path with a mobile robot - Google Patents

Abstract

In a method according to the invention for traversing a path with a mobile robot (1), a predefined starting path (10) is adapted with respect to a desired path (20; 20'; 20'') on the basis of a quality function which depends both on a clearance of the robot on the desired path and on a difference between the starting path and the desired path.

Description

description

Driving a path with a mobile robot

The present invention relates to a method and system for following a path with a mobile robot and a computer program product for

Implementation of the procedure.

Mobile robots should often follow predefined paths, for example to monitor rooms, transport goods or the like. In particular with the simultaneous determination of poses and map creation (“Simultaneous Localization and Mapping”, SLAM), the initial specification of a path for exploring an environment can be useful.

The object of the present invention is to improve the ability to follow a path with a mobile robot.

This object is achieved by a method with the features of claim 1. Claims 9, 10 provide a system or computer program product

Carrying out a procedure described here under protection. The subclaims relate to advantageous developments.

According to one embodiment of the present invention, in particular, a

Output path specified for a mobile robot. According to one embodiment of the present invention, this output path is adapted to a target path which the robot is to follow in one embodiment (instead of the original or non-adapted output path).

In one embodiment, the mobile robot has a mobile base and / or a robot arm which in one embodiment has at least three, in particular at least six, in one embodiment at least seven joints or (movement) axes. Additionally or alternatively, the mobile robot is an autonomous robot in one embodiment. In the case of such mobile robots, the present invention can be used with particular advantage because of its diverse application possibilities. In one embodiment, the starting path and the target path each have nodes and edges connecting them. In this way, paths can advantageously, in particular efficiently, be specified and / or edited or processed in one embodiment.

Through an (initial) specification of an exit path, as explained in the introduction, in particular a simultaneous determination of the pose and map creation (SLAM) can be improved, so that in one embodiment the robot creates a map of the surroundings while trying to find the (adapted to the target path) Follow the exit path. In this case, creating a map can in particular also include changing an already existing map.

However, the present invention is not limited to this, but can be used with particular advantage in order to follow an output path adapted to the target path, multiple times in one embodiment, with the mobile robot, in particular for exploration, monitoring and / or transport purposes or similar.

According to one embodiment of the present invention, the predetermined

Output path based on, in particular with optimization, a quality function, in an embodiment with minimization of a cost function or analogously with maximization of a utility function, adapted to the target path, both from a free space of the robot on the target path and from a difference between the

Output path and target path depends, in particular on the basis of the free space and the difference is determined, iteratively in one embodiment.

In one embodiment, the quality function depends on the free space and the

Difference from the fact that with the same difference it becomes better with increasing free space, in particular (as a cost function) smaller or (as a utility function) larger, and / or with the same free space with increasing difference it becomes worse, in particular (as a cost function) larger or (than Utility function) is smaller, is, or is determined in this way or is, in particular is defined in this way, in an embodiment as

weighted sum g = g-, + g ₂ or the like.

In this way, in one embodiment, a predetermined output path can advantageously be followed as precisely as possible and, in the process, a collision of the mobile robot his environment should be avoided. As a result, in one embodiment, the specified output path or advance information can advantageously be used to perform an exploration task, in particular a simultaneous determination of poses and map creation, a monitoring and / or a transport task better, in particular more efficiently and / or more reliably.

In one embodiment, the starting path is online or while the robot is trying, in particular for the first time, to travel along the target path, adapted to the target path, in one embodiment iteratively. In a further development, the robot tries, in one embodiment, to approach a respective next node of the path with the aid of dynamic path planning. In one embodiment, the starting path, in particular here, is adapted step-by-step or successively to the target path, in one embodiment each time a node is reached.

As a result, in one embodiment, an exploration task, in particular a simultaneous determination of a pose and map creation, a monitoring and / or a transport task can be carried out better, in particular more efficiently and / or more reliably.

In one embodiment, the free space of the robot on the target path or a corresponding factor of the quality function depends on one or more minimal ones

Distances of the robot, in particular, cumulatively in one embodiment, of

minimum distances of the robot at nodes and / or minimum distances of the robot to edges of the target path connecting these nodes to an environment, in particular obstacles in the environment, and / or a predetermined one

Target free space, which is greater than zero in one embodiment, in one embodiment such that the quality function better with increasing minimum distances, in particular (as a cost function) smaller or (as a utility function) larger, or the free space or corresponding Factor of the quality function becomes larger with increasing minimum distances, in a further development for minimum distances that are larger than the specified target free space remains the same, for example in the form: ίq <=> d {xi)> target free space

QM =

l Desired free space - d (Xj) <=> otherwise 9i = 9u with the minimum distance d at the point x, in particular a node or a point on an edge with a minimum distance from the surroundings, or is determined in this way. The free space of the robot can in particular be a free mobility of the robot with a predetermined or any orientation.

In one embodiment, the free space of the robot on the target path is prioritized when adapting the output path to the target path over the difference between the output path and target path, in particular such that one or more minimal distances between the robot and its surroundings are always is greater than zero.

In one embodiment, this can advantageously reduce the risk of a collision.

In one embodiment, the free space of the robot is determined on the basis of a map of the surroundings of the robot, in one embodiment on the basis of a map of the surroundings that the robot creates while it is trying to travel the target path, and / or on the basis of a distance map, which in turn based on the area map and for

different places each indicate the minimum distance to the environment.

As a result, the free space can advantageously, in particular efficiently and / or quickly, be determined in one embodiment.

In particular, if the starting path is adapted to the target path while the robot is trying to follow the target path, only the part of the target path that the robot has (already) tried to follow is taken into account for the free space in one embodiment.

As a result, in one embodiment, the output path can advantageously be adapted to the target path, in particular successively or iteratively and / or on the basis of an environment that has already been explored.

In one embodiment, the difference between the initial path and the target path or a corresponding factor of the quality function depends on a length difference, Shape difference and / or position difference between the output path and target path, in an embodiment such that the quality function

- With the same shape and / or pose difference, the longer the length difference between the initial path and the target path becomes worse, in particular larger (as a cost function) or smaller (as a utility function);

- With the same length and / or position difference, the difference in shape between the starting path and the target path becomes worse, in particular larger (as a cost function) or smaller (as a utility function); and or

- With the same length and / or shape difference, the positional difference between the starting path and the target path increases with increasing position, in particular (as a cost function) larger or (as a utility function) smaller;

in particular the difference between the initial path and the target path or the corresponding factor of the quality function

- With the same shape and / or position difference, the length difference between the starting path and the target path increases with increasing length;

- With the same length and / or positional difference, the shape difference between the starting path and the target path increases with increasing shape; and or

- With the same length and / or shape difference, the positional difference between the starting path and the target path increases;

or is determined in this way.

In one embodiment, the shape difference or the corresponding factor of the quality function, in one embodiment cumulatively, is based on angles of the output path connecting nodes of the output path and edges of the target path connecting angles between nodes of the target path, in particular a difference between Angles of the output path and (corresponding) angles of the target path are determined, in one embodiment such that the shape difference or the corresponding factor of the quality function increases with increasing differences between angles of the output path and (corresponding) angles of the target path or the quality function gets worse.

Additionally or alternatively, the position difference or the corresponding factor of the quality function is determined in one embodiment on the basis of distances between the nodes of the output path and (corresponding) nodes of the target path, in a Execution in such a way that the position difference or the corresponding factor of the quality function increases or the quality function becomes worse with increasing distances between nodes of the output path and (corresponding) nodes of the target path.

In a further development, the distances are or are weighted with a distance to a (specific, in one embodiment the first or start) node. As a result, a rotation around this specific node can advantageously be taken into account in one embodiment, since the resulting displacements of the nodes that increase with the distance from this specific node have less of an influence on the quality function.

Additionally or alternatively, in one embodiment, the difference between the output path and the target path depends on a scaling factor between the

Starting path and target path, in one development the length and / or position difference is weighted differently on the basis of this scaling factor, weighted less in one embodiment if a scaling factor between the

Starting path and target path deviate from one. As a result, it can advantageously be taken into account in one embodiment that a scaling error between the output path and the target path can lead to an incorrect length or position difference.

In this way, in each case, in particular in combination of two or more of the aforementioned features, the output path can be maintained as well as possible.

According to one embodiment of the present invention, a system, in particular hardware and / or software, in particular in terms of program technology, is set up to carry out a method described here and / or has: Means for adapting a predetermined output path to a target path on the basis of a quality function , wherein the quality function depends both on a free space of the robot on the target path and on a difference between the starting path and target path.

In one embodiment, the system or its means has:

Means for adapting the exit path to the target path while the robot attempts to travel the target path; and or Means for determining the free space of the robot on the basis of a map of the surroundings of the robot, in particular means for creating a map of the surroundings while the robot is trying to follow the target path; and or

Means for determining the shape difference on the basis of angles between edges of the output path connecting nodes of the path and the desired path and / or to the

Determining the position difference on the basis of, in particular weighted with a distance to a node, distances between nodes of the output path and target path; and or

Means for, in particular multiple, traversing the output path adapted to the target path with or by the robot.

A means within the meaning of the present invention can be designed in terms of hardware and / or software, in particular a processing, in particular microprocessor unit (CPU), graphics card (GPU), preferably a data or signal connected to a memory and / or bus system, in particular a digital processing unit ) or the like, and / or one or more programs or program modules. The

Processing unit can be designed to receive commands as one in one

Program stored in the storage system are implemented,

Acquire input signals from a data bus and / or deliver output signals to a data bus. A storage system can have one or more,

in particular have different storage media, in particular optical, magnetic, solid-state and / or other non-volatile media. The program can be designed in such a way that it embodies or is able to execute the methods described here, so that the processing unit carries out the steps of such

Can perform method and thus in particular can control the mobile robot. In one embodiment, a computer program product can have, in particular, a non-volatile storage medium for storing a program or with a program stored thereon, execution of this program causing a system or a controller, in particular a computer, to create a program here to carry out the described method or one or more of its steps.

In one embodiment, one or more, in particular all, steps of the

The method is carried out completely or partially automatically, in particular by the system or its means. In one embodiment, the system has the mobile robot.

Further advantages and features emerge from the subclaims and the exemplary embodiments. This shows, partly schematically:

Fig. 1 - 5: a mobile robot when following a path after a

Carrying out the present invention.

Fig. 2 -» Fig. 3 -» Fig. 4 -» Fig. 5 zeigt einen mobilen Roboter 1 beim Abfahren eines Pfades nach einer Ausführung der vorliegenden Erfindung und damit gleichermaßen ein Verfahren wie auch ein System nach einer Ausführung der vorliegenden Erfindung. The sequence of figures Fig. 1

Fig. 2 - »Fig. 3 -» Fig. 4 - »Fig. 5 shows a mobile robot 1 traveling along a path according to an embodiment of the present invention and thus a method as well as a system according to an embodiment of the present invention.

The robot 1 is located in an environment that is indicated by two obstacles 60, 61.

For the simultaneous determination of the poses and the creation of a map, an output path 10 is given, the nodes, of which only some nodes 11, 12, 13 are provided with reference symbols for a better overview, and edges connecting nodes, of which only one edge 14 is provided with reference symbols for a better overview, having.

In FIG. 1, starting from starting node 11, the robot 1 tries to approach the next node of the starting path 10 with the aid of dynamic path planning. However, this would be in the obstacle 60, so that the robot cannot approach this node.

Accordingly, it adapts the output path 10 to a target path 20. To this end, he minimizes a quality or cost function, which is a weighted sum of a free space of the robot 1 on the target path and a difference between the

Output path 10 and the target path 20 includes, in one embodiment.

The consideration of the free space compared to the obstacle 60 detected when driving down as well as the specification, length, shape and poses of the

to maintain the specified output path as far as possible, in

Exemplary embodiment for the rotation of the output path 10 in FIG. 2 in the target path 20. The robot only takes into account the free space on the part of the path that it has tried to follow up to now.

The robot 1 now tries to approach the next node of the output path 10 adapted to the target path 20 with the aid of dynamic path planning. However, this would again be in the obstacle 60, so that the robot cannot approach this node.

Accordingly, in its pose shown in FIG. 2, it adapts the output path 10 adapted to the target path 20 further to a target path 20 shown in FIG. 3 and minimizes the quality or cost function for this. Again, the robot only takes into account the free space on the part of the path that it has tried to follow up to now.

This leads to the further rotation, which can be seen in FIG. 3, of the output path 10 adapted to the desired path 20 in the desired path 20 ‘shown in FIG. 3.

The robot 1 now tries successively to approach the next nodes of the output path 10 adapted to the target path 20 'with the aid of dynamic path planning.

In doing so, however, it would collide with obstacle 61.

Accordingly, in the further course of the process it adapts the output path 10 adapted to the target path 20 ‘further to the target path 20 ″ shown in FIG. 4 and minimizes the quality or cost function for this.

This leads to the further rotation, recognizable in FIG. 4, of the output path 10, which is adapted to the desired path 20 Pfad, in the desired path 20 ″ shown in FIG. 4.

On the one hand, this has sufficient free space everywhere to the obstacles 60, 61 and, on the other hand, corresponds as far as possible, in particular with regard to its length, shape and pose, to the initially specified exit path 10. Correspondingly, the robot 1 now travels on this output path 10, which has been adapted to the target path 20 ″, or travels again, if necessary several times.

The sequence of figures Fig. 1 - Fig. 2 Fig. 3 -> Fig. 4 - »Fig. 5 makes it clear that the robot 1 successively the starting path 10 to the target path 20 ->

20 ”and thereby minimizing the quality or cost function described above and trying to follow the adapted output path, as indicated by the actually driven path 2.

Although exemplary embodiments have been explained in the preceding description, it should be pointed out that a large number of modifications are possible. It should also be noted that the exemplary

Explanations are only examples that cover the scope of protection that

Applications and structure are not intended to restrict in any way. Rather, the person skilled in the art through the preceding description is a guide for

Implementation of at least one exemplary embodiment given, whereby various changes, in particular with regard to the function and arrangement of the described components, can be made without departing from the scope of protection as it emerges from the claims and their equivalents

Combinations of characteristics results.

Reference list

1 mobile robot

2 crazy way

10 default output path

1 1-13 knots

14 edge

20; 20 ‘; 20 “target path

60; 61 obstacle

Claims

Claims 1. A method for following a path with a mobile robot (1), wherein a predetermined output path (10) is adapted on the basis of a quality function to a target path (20; 20 '; 20 "), which is both from a free space of the robot depends on the target path as well as on a difference between the initial path and target path. 2. The method according to claim 1, characterized in that the starting path is adapted to the target path, while the robot tries to travel the target path. 3. The method according to any one of the preceding claims, characterized in that the output path and the target path each have nodes (11-13) and edges (14) connecting them. 4. The method according to any one of the preceding claims, characterized in that the free space of the robot on the target path of at least a minimum distance of the robot, in particular at nodes and / or edges of the target path connecting them, to an environment and / or depends on a predetermined target clearance. 5. The method according to any one of the preceding claims, characterized in that the free space of the robot is determined on the basis of a map of the surroundings of the robot, in particular a map of the surroundings that the robot creates while it is trying to travel the target path, and / or that only that part of the target path that the robot has tried to follow is taken into account for the free space. 6. The method according to any one of the preceding claims, characterized in that the difference between the output path and target path of one Length difference, shape difference, position difference and / or a scaling factor between the output path and the target path depends. 7. The method according to the preceding claim, characterized in that the shape difference based on angles between nodes (1 1-13) of the path connecting edges (14) of the output path (10) and target path (20; 20 '; 20 " ) is determined and / or the positional difference on the basis of, in particular with a distance to a node (11) weighted, distances between nodes (12, 13) of the output path (10) and the target path (20; 20 '; 20 ") is determined. 8. The method according to any one of the preceding claims, characterized in that the robot uses the output path adapted to the target path, especially multiple times. 9. System for following a path with a mobile robot (1) that is used to Implementation of a method according to one of the preceding claims is set up and / or means for adapting a predetermined one Output path (10) on the basis of a quality function for a desired path (20; 20 ‘; 20 ″) which depends both on a free space of the robot on the desired path and on a difference between the output path and the desired path. 10. Computer program product with a program code, which is stored on a medium readable by a computer, for performing a method according to one of the preceding claims.