Automatic following luggage case control method

Abstract

The invention provides an automatic following luggage case control method, and belongs to the field of positioning and following. Obstacles in the application environment of the automatic following luggage case are dense, and if the obstacles suddenly appear or move, a wrong target or a lost target can be easily followed. According to the automatic following luggage case control method provided by the invention, a plurality of signal receiving devices and wireless sensors are uniformly arranged on the periphery of the automatic following luggage case, a path is planned by adopting a composite path planning strategy fusing space division and an A * algorithm according to received signals and environmental parameters, the following speed and angle are adjusted in real time, and the automatic following luggage case is controlled. The following real-time performance and accuracy of the automatic following luggage case are improved, and the activity range of the automatic following luggage case is expanded.

Description

Technical Field

[0001] The present invention relates to the field of positioning and following technology, and in particular to a control method for automatically following a luggage case. Background Art

[0002] Trolley suitcases are essential tools for long-distance travel, allowing people to carry a lot of items, but people's hands are still not free, and long-term dragging makes people tired. Automatic following suitcases came into being, which can follow the owner autonomously and free the owner's hands. When the automatic following suitcase follows the owner, it first accurately locates the owner, then plans the path, and finally performs following control.

[0003] The main positioning and following path planning strategies are Dijkstra algorithm, A* algorithm and RRT algorithm. Dijkstra algorithm is a classic shortest path algorithm that can find the shortest path from a starting point to all other nodes. A* algorithm is a heuristic search algorithm that combines the best-first search of Dijkstra algorithm with a heuristic function (usually Euclidean distance or Manhattan distance) to find the shortest path from the starting point to the end point. The A* algorithm relies on the quality of the heuristic function. If the heuristic function is not accurate enough, it may affect the efficiency and quality of path planning. RRT algorithm is a random sampling method for finding a path from the starting point to the end point in a high-dimensional space. It explores the space by continuously expanding the tree structure. The RRT algorithm requires a large amount of sampling to find a feasible path, and the path found may not be the optimal path.

[0004] The application environment of the automatic luggage tracking system is densely populated with obstacles. If an obstacle suddenly appears or moves, it is easy to mis-track or lose the target. Therefore, it is urgent to provide a solution to improve this problem. Summary of the invention

[0005] The object of the present invention is to provide a control method for an automatic following suitcase, which can solve the problem that the automatic following suitcase is prone to following the wrong target or losing the target.

[0006] The present invention provides a method for controlling an automatic following luggage case, comprising the following technical solutions:

[0007] Obtain the signal of the target to be followed and automatically follow the environmental parameters around the suitcase;

[0008] Calculating the relative position of the following target and the automatically following luggage box based on the signal and the environmental parameters;

[0009] Based on the relative position, a composite path planning strategy integrating space segmentation and A* algorithm is used to plan a path for automatically following the suitcase;

[0010] The following speed and angle of the automatically following suitcase are adjusted according to the planned path.

[0011] The present invention provides a control method for an automatic following suitcase. A plurality of signal receiving devices and wireless sensors are evenly arranged around the automatic following suitcase. A composite path planning strategy integrating space segmentation and A* algorithm is adopted to plan a path according to received signals and environmental parameters. The following speed and angle are adjusted in real time, thereby improving the real-time and accuracy of the automatic following suitcase and expanding the activity range of the automatic following suitcase.

[0012] Optionally, when acquiring the signal of the following target and the environmental parameters around the automatically following suitcase, a plurality of signal receiving devices and wireless sensors are evenly arranged around the automatically following suitcase.

[0013] Optionally, when calculating the relative position of the following target and the automatically following suitcase based on the signal and environmental parameters, the normalized signal strength and the signal strength nonlinear correction value are calculated respectively according to the signal strength, and then the single-point distance is calculated respectively according to the signal strength, the environmental parameters and the signal strength nonlinear correction value, and finally, the relative position of the following target and the automatically following suitcase is calculated by combining multiple single-point distances and normalized signal strengths.

[0014] Optionally, when the relative position is calculated by integrating the multiple single-point distances and normalized signal strengths, a spatial coordinate system is constructed based on the middle position of the initial state of the following device as the coordinate origin, and the position coordinates of the following target in the spatial coordinate system are calculated.

[0015] Optionally, when planning the path of the automatic following suitcase based on the relative position using a composite path planning strategy that integrates spatial segmentation and the A* algorithm, the strategy first uses spatial segmentation technology to divide the environment into multiple sub-areas, and uses the initial position of the following device as the starting node, and the relative position of the following target as the target node, and then independently applies the A* algorithm in each sub-area to search for a local optimal path, and connects each local path through boundary nodes to form a global shortest path from the starting node to the target node.

[0016] Optionally, when searching for the global shortest path based on a composite path planning strategy that integrates spatial segmentation and the A* algorithm, not only the Dijkstra distance of the node is considered, but also a dynamic weight evaluation mechanism is introduced to calculate the comprehensive cost of each adjacent node to the target node. The dynamic weight evaluation mechanism dynamically adjusts the weight according to factors such as environmental complexity, obstacle density, and path smoothness.

[0017] Optionally, when adjusting the following speed and angle of the automatically following suitcase according to the planned path, the deviation angle between the following device along the planned path and the following target is obtained in real time, and the following speed and following angle of the following device are dynamically adjusted based on the deviation angle and a preset control strategy of an adaptive PID controller, and the proportional, integral and differential gain coefficients in the control strategy of the adaptive PID controller are adaptively adjusted according to the motion characteristics and environmental conditions of the following device. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 The present invention provides a flowchart of a method for automatically following a luggage compartment. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. Unless otherwise defined, the technical terms or scientific terms used herein should be understood by people with general skills in the field to which the present invention belongs. "Including" and similar words used in this article mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects.

[0020] An embodiment of the present invention provides a method for controlling an automatic following luggage case, comprising the following steps:

[0021] S1, obtaining the signal of the following target and the environmental parameters around the automatically following suitcase;

[0022] S2. Calculating the relative position of the following target and the automatically following suitcase based on the signal and environmental parameters;

[0023] S3, planning a path for automatically following the suitcase using a composite path planning strategy integrating space segmentation and A* algorithm based on the relative position;

[0024] S4. Adjusting the following speed and angle of the automatically following suitcase according to the planned path.

[0025] In fact, when executing step S1,

[0026] Multiple signal receiving devices are evenly arranged around the automatic following suitcase, so that the recognition accuracy of the receiving device facing the following targets in different directions can be effectively improved, and the following device can have multiple working surfaces with the same effect without adjusting the direction of the following device. In some embodiments, the receiving device can be a Bluetooth device, and the received tag signal can be a Bluetooth signal.

[0027] Multiple wireless sensors are evenly arranged around the automatic following suitcase. These wireless sensors are cleverly embedded in various key parts of the automatic following suitcase, including but not limited to its top, bottom, sides and other possible locations, to ensure that multi-dimensional information of the surrounding environment can be fully and accurately collected. This information includes temperature, humidity, obstacles such as walls, and light intensity.

[0028] By combining data from multiple receiving devices and wireless sensors, the signal coverage is expanded and the effective following distance is extended.

[0029] In some embodiments, when executing step S1, the target to be followed may be a person, an animal, or other movable object. Specifically, the moving speed of the target to be followed is less than or equal to 5 m / s. Specifically, the target to be followed may be a device worn on the surface of a person or an animal, such as a bracelet, a ring, or other electronic device that can transmit a signal.

[0030] In fact, when executing step S2, the normalized signal strength and the signal strength nonlinear correction value are calculated according to the signal strength of the tag signal received by the signal receiving device, and then the single-point distance is calculated according to the signal strength, the environmental parameters collected by the wireless sensor and the signal strength nonlinear correction value, and finally the relative position of the following target and the automatic following suitcase is calculated by combining multiple single-point distances and normalized signal strengths. The normalized signal strength and the nonlinear correction value improve the target recognition accuracy and reduce the situation of following the wrong target. The introduction of environmental parameter correction further reduces the possibility of misidentification, so that the automatic following suitcase can accurately identify the target even in a complex environment.

[0031] S21, when respectively calculating the normalized signal strength and the signal strength nonlinear correction value based on the signal strengths of the plurality of tag signals, the calculation formula is as follows:

[0032]

[0033]

[0034] in: is the normalized signal strength between the i-th receiving device and the tracking target; RSSI i RSSI is the signal strength of the tag signal between the i-th receiving device and the tracking target; MINThe minimum signal strength of the tag signal between all receiving devices and the target being followed; RSSI MAX The maximum signal strength of the tag signal between all receiving devices and the tracking target; f(RSSI i ) is the nonlinear correction value of the signal strength of the tag signal between the i-th receiving device and the followed target; a and b are the optimal fitting parameters.

[0035] S22, when calculating the single point distance according to the signal strength, the environmental parameters collected by the wireless sensor and the signal strength nonlinear correction value, the calculation formula is as follows:

[0036]

[0037] Where: d i is the single-point distance between the i-th receiving device and the tracking target; RSSI i is the signal strength of the tag signal between the i-th receiving device and the tracking target; f(RSSI i ) is the nonlinear correction value of the signal strength of the tag signal between the i-th receiving device and the tracking target; Δenv is the environmental correction term obtained by predicting the environmental parameters through the machine learning model; A i is the preset calibration coefficient of the i-th receiving device; n is the environmental signal attenuation factor.

[0038] S23. When calculating the relative position of the following target and the automatic following suitcase by integrating the multiple single-point distances and the normalized signal strengths, a spatial coordinate system is constructed based on the middle position of the initial state of the following device as the coordinate origin, and the position coordinates of the following target in the spatial coordinate system are calculated.

[0039] Specifically, when calculating the position coordinates of the follow-up target in the space coordinate system, the calculation formula is as follows:

[0040]

[0041]

[0042]

[0043] Where: d i is the single-point distance between the i-th receiving device and the tracking target; is the normalized signal strength between the i-th receiving device and the target being followed; (X, Y, Z) is the position coordinate of the target being followed in the spatial coordinate system; (X i ,Y i ,Z i ) is the current position coordinate of the i-th receiving device in the spatial coordinate system.

[0044] In fact, when executing step S3, a composite path planning strategy integrating space segmentation and A* algorithm is adopted. This strategy first uses space segmentation technology to divide the environment into multiple sub-areas, and uses the initial position of the automatic following suitcase as the starting node, and the relative position of the following target as the target node. Subsequently, the A* algorithm is independently applied in each sub-area to search for the local optimal path, and each local path is connected through the boundary node to form the global shortest path from the starting node to the target node. Through this composite path planning strategy, the automatic following suitcase can work stably during the following process, while expanding the following range.

[0045] Specifically, when searching for the shortest path from the starting node to the target node based on the A* algorithm, not only the Dijkstra distance of the node is considered, but also a dynamic weight evaluation mechanism is introduced to calculate the comprehensive cost of each adjacent node to the target node. This mechanism dynamically adjusts the weight according to factors such as environmental complexity, obstacle density, and path smoothness, thereby ensuring that the path length and path safety can be balanced during the search process.

[0046] S31. When calculating the comprehensive cost from each adjacent node to the target node, the calculation formula is as follows:

[0047] C(i)=W 1 *D(i)+W 2 *H(i)+W 3 *S(i)

[0048] Where: C(i) is the comprehensive cost from the i-th node to the target node; D(i) is the Dijkstra distance from the i-th node to the target node; H(i) is the estimated cost from the i-th node to the target node, calculated using the Euclidean distance or Manhattan distance; S(i) is the smoothness evaluation value of the path segment where the i-th node is located; W 1 , W 2 , W 3 are the weight coefficients of the corresponding items, which are determined by the dynamic weight evaluation mechanism.

[0049] S32, when calculating the estimated cost from each adjacent node to the target node and the smoothness evaluation value of the path segment where each adjacent node is located, the Euclidean distance is used as a heuristic function, and the heuristic function is used to evaluate the estimated cost from the current position to the target position. The calculation formula is as follows:

[0050]

[0051]

[0052] Where: h(i) is the estimated cost from the i-th node to the target node; (X i , Y i, Z i ) is the current position coordinate of the i-th node in the spatial coordinate system; (X, Y, Z) is the position coordinate of the target node in the spatial coordinate system; S(i) is the smoothness evaluation value of the path segment where the i-th node is located; L(i) is the length of the line segment between the i-th node and its adjacent nodes.

[0053] Furthermore, in order to verify the superiority of the composite path planning strategy integrating spatial segmentation and A* algorithm, in the Python environment, the Dijkstra algorithm, A* algorithm, RRT algorithm and the composite path planning strategy integrating spatial segmentation and A* algorithm were used to set fixed starting points and end points on randomly generated maps of three sizes to obtain evaluation indicators. The average results obtained by performing 30 experiments are shown in the following table:

[0054]

[0055] Among them, three scales of maps are randomly generated using Python and Matplotlib: 1. Small-scale map: 10x10 grid, containing 10% obstacles.

[0056] 2. Medium-sized map: 50x50 grid, containing 20% ​​obstacles.

[0057] 3. Large-scale map: 100x100 grid, containing 30% obstacles.

[0058] The evaluation index is set as:

[0059] 1. Path length: The actual path length from the starting point to the end point (unit: grid number).

[0060] 2. Computation time: the time required for path planning (unit: milliseconds).

[0061] 3. Path smoothness: The smoothness of the path is measured by the change in path curvature (the inverse of the standard deviation of curvature, unit: radians / grid).

[0062] 4. Path safety: The ability of a path to avoid obstacles, measured by the closest distance between the path and the obstacle (unit: number of grids)

[0063] From the comparison of each evaluation index in the table, it can be seen that

[0064] 1. Path length: At all map scales, the path length of the fusion of spatial segmentation and A* algorithm is similar to that of the A* algorithm and is better than other algorithms; on large-scale maps, the path length of the fusion of spatial segmentation and A* algorithm is slightly better than other algorithms.

[0065] 2. Computation time: On small and medium-scale maps, the computation time of the fusion of spatial segmentation and A* algorithm is slightly higher than that of the A* algorithm, but the difference is not significant. However, on large-scale maps, the computation time of the fusion of spatial segmentation and A* algorithm is significantly lower than that of other algorithms, showing its advantages in large-scale environments.

[0066] 3. Path smoothness and path safety: The path smoothness of the fusion of spatial segmentation and A* algorithm is the highest, which is better than other algorithms. At the same time, the path safety is the highest, and the closest distance between the path and the obstacle is the largest.

[0067] Therefore, the composite path planning strategy that integrates spatial segmentation and the A* algorithm performs well in terms of path planning efficiency and path quality. This strategy is particularly suitable for path planning in large-scale and complex environments, and can provide more efficient and safer path planning solutions.

[0068] In fact, when executing step S4, the following speed and angle of the automatically following suitcase are adjusted according to the planned path, the deviation angle between the following device and the following target along the global shortest path is obtained in real time, and the following speed and following angle of the following device are dynamically adjusted based on the deviation angle and the control strategy of the preset adaptive PID controller. The proportional, integral and differential gain coefficients in the control strategy of the adaptive PID controller are adaptively adjusted according to the motion characteristics and environmental conditions of the following device to achieve more accurate and stable following control. Through the adaptive PID controller, the automatically following suitcase can respond to the user's movement in a timely manner, and can continue to stably track the target even in complex environmental conditions such as obstacle obstruction or signal attenuation, reducing the situation of losing the target and improving the robustness of the following.

[0069] S41. When obtaining the deviation angle, the calculation formula is as follows:

[0070] α=tan2(x a -x b ,y a -y b );

[0071] Where: α is the deviation angle of the following device; (x a ,y a ) is the relative coordinate of the target in the XY plane; (x b ,y b ) is the relative coordinate of the current node in the XY plane;

[0072] S42, when controlling the following speed and following angle of the following device based on the deviation angle, the calculation formula is as follows:

[0073]

[0074] θ=arctan2(x a -x b ,y a -y b ,);

[0075] Where: V is the following speed of the following device; k p k is the proportional gain coefficient of the PID controller in the following device; i k is the integral gain coefficient of the PID controller in the following device; d is the differential gain coefficient of the PID controller in the following device; θ is the direction angle of the following device.

[0076] Although the embodiments of the present invention are described in detail above, it is obvious to those skilled in the art that various modifications and variations can be made to these embodiments. However, it should be understood that such modifications and variations are within the scope and spirit of the present invention as described in the claims. Moreover, the present invention described herein may have other embodiments and may be implemented or realized in a variety of ways.

Claims

1. A method for automatically following a luggage compartment, characterized in that: Obtain the signal of the target to be followed and automatically follow the environmental parameters around the suitcase; Calculating the relative position of the following target and the automatically following luggage box based on the signal and the environmental parameters; Based on the relative position, a composite path planning strategy integrating space segmentation and A* algorithm is used to plan a path for automatically following the suitcase; The following speed and angle of the automatically following suitcase are adjusted according to the planned path.

2. The automatic following luggage control method according to claim 1, characterized in that: When acquiring the signal of the following target and the environmental parameters around the automatic following suitcase, a plurality of signal receiving devices and wireless sensors are evenly arranged around the automatic following suitcase.

3. The automatic following luggage control method according to claim 1, characterized in that: When calculating the relative position of the following target and the automatically following suitcase based on the signal and environmental parameters, the normalized signal strength and the signal strength nonlinear correction value are calculated according to the signal strength, and then the single-point distance is calculated according to the signal strength, the environmental parameters and the signal strength nonlinear correction value, and finally the relative position of the following target and the automatically following suitcase is calculated by combining multiple single-point distances and normalized signal strengths.

4. The automatic following luggage control method according to claim 3, characterized in that: When the relative position is calculated by integrating the multiple single-point distances and normalized signal strengths, a spatial coordinate system is constructed based on the middle position of the initial state of the following device as the coordinate origin, and the position coordinates of the following target in the spatial coordinate system are calculated.

5. The automatic following luggage control method according to claim 1, characterized in that: When planning the path of the automatic following suitcase based on the relative position using a composite path planning strategy that integrates space segmentation and the A* algorithm, the strategy first uses space segmentation technology to divide the environment into multiple sub-areas, and uses the initial position of the following device as the starting node, and the relative position of the following target as the target node, then independently applies the A* algorithm in each sub-area to search for the local optimal path, and connects the local paths through boundary nodes to form a global shortest path from the starting node to the target node.

6. The automatic following luggage control method according to claim 5, characterized in that: When searching for the global shortest path based on the composite path planning strategy integrating spatial segmentation and A* algorithm, not only the Dijkstra distance of the node is considered, but also a dynamic weight evaluation mechanism is introduced to calculate the comprehensive cost of each adjacent node to the target node. The dynamic weight evaluation mechanism dynamically adjusts the weight according to factors such as environmental complexity, obstacle density and path smoothness.

7. The automatic following luggage control method according to claim 1, characterized in that: When adjusting the following speed and angle of the automatically following suitcase according to the planned path, the deviation angle between the following device and the following target along the planned path is obtained in real time, and the following speed and following angle of the following device are dynamically adjusted based on the deviation angle and the control strategy of the preset adaptive PID controller, and the proportional, integral and differential gain coefficients in the control strategy of the PID controller are adaptively adjusted according to the motion characteristics and environmental conditions of the following device.

Images