Intelligent wheelchair local path planning and dynamic obstacle avoidance method based on rolling window

Abstract

The invention relates to the technical field of mobile equipment, in particular to an intelligent wheelchair local path planning and dynamic obstacle avoidance method based on a rolling window. The method comprises the following steps: S1, establishing an indoor local three-dimensional map based on a binocular vision sensor; s2, determining a local optimal path according to the established indoor local three-dimensional map; s3, boundary information and position information of the obstacle are obtained through a front infrared / binocular vision sensor, the position of the obstacle relative to the wheelchair is obtained, the rotating speed is distributed to left and right wheels of the wheelchair through a differential, and arc obstacle avoidance is achieved in combination with a steering engine. According to the rolling window-based intelligent wheelchair local path planning and dynamic obstacle avoidance method provided by the invention, an indoor instantaneous local map is created based on binocular vision, a local path is planned based on the rolling window, full-autonomous trajectory planning of the intelligent wheelchair is realized, and based on an infrared / binocular vision sensor, various environmental limitations are considered at the same time, and the dynamic obstacle avoidance of the intelligent wheelchair is realized. A dynamic obstacle avoidance method is designed, and the obstacle avoidance effect of the machine in a dynamic environment is improved.

Description

Technical Field

[0001] This invention relates to the field of mobile device technology, and in particular to a method for local path planning and dynamic obstacle avoidance of intelligent wheelchairs based on a scrolling window. Background Technology

[0002] With the increasing aging of the population and the gradual decline in physical function with age, more and more countries are strengthening research on intelligent wheelchairs to address the mobility challenges faced by the elderly. In today's era of increasingly prevalent artificial intelligence technology, intelligent wheelchairs require improved autonomous movement capabilities, environmental perception, and intelligent decision-making abilities. Addressing the instability and safety issues of current intelligent wheelchairs' autonomous navigation in unknown environments, this paper proposes a local path planning and dynamic obstacle avoidance method based on a scrolling window. This aims to overcome the shortcomings of current intelligent wheelchairs in complex indoor environments, such as their inability to fully autonomously plan paths and poor obstacle avoidance performance. This research is of great significance for improving the mobility and quality of life of the elderly in their daily lives. Summary of the Invention

[0003] Therefore, it is necessary to provide a local path planning and dynamic obstacle avoidance method for intelligent wheelchairs based on a scrolling window to address the aforementioned technical problems.

[0004] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: The method for local path planning and dynamic obstacle avoidance of intelligent wheelchairs based on scrolling windows has the following steps: S1: Establishing a local 3D map of the indoor area based on a binocular vision sensor; S2: Determine the local optimal path based on the established indoor local 3D map; S3: Obtains the boundary and position information of obstacles through a front-facing infrared / binocular vision sensor, obtains the position of the obstacle relative to the wheelchair, distributes the rotation speed to the left and right wheels of the wheelchair through a differential, and achieves circular obstacle avoidance in combination with the servo motor.

[0005] As a preferred embodiment of the intelligent wheelchair local path planning and dynamic obstacle avoidance method based on a scrolling window provided by the present invention, in step S2, the entire map is traversed according to the indoor local three-dimensional map to determine the location of all exits, each exit is connected to the endpoint, the shortest distance is used as the evaluation criterion to determine the optimal solution, and it is used as the virtual target point of the local map.

[0006] As a preferred embodiment of the intelligent wheelchair local path planning and dynamic obstacle avoidance method based on a scrolling window provided by the present invention, the virtual target point is used as the basis to plan the optimal path under the visualization map. This process is repeated until the destination is reached. If there is no exit, the wheelchair is designed to return to the previous target point along the original route, and the suboptimal solution is selected for path planning.

[0007] As a preferred embodiment of the intelligent wheelchair local path planning and dynamic obstacle avoidance method based on a scrolling window provided by the present invention, the virtual target point is used as the shortest distance as the evaluation criterion, and the steps are as follows: A. When the wheelchair is in position 1, select exit 1 and exit 2 as seen within the field of vision; B. Connect position 1 to exit 1 and exit 1 to the end point respectively to form Line 1. Connect position 1 to exit 2 and exit 2 to the end point respectively to form Line 2. C. Determine the distance between the two lines in step B, confirm the optimal path, and use it as the virtual target point within this visualization range; D. When the wheelchair moves to position 2, select exit 3 and exit 4 as seen within the field of vision; E. Connecting position 2 with exit 3, and exit 3 with the end point, forms Line 3; connecting position 2 with exit 4, and exit 4 with the end point, forms Line 4. F. Determine the distance between the two lines in step E, confirm the optimal path, and use it as the virtual target point within this visualization range; G. Move the wheelchair to the new position and repeat steps AF until the destination is reached.

[0008] As a preferred embodiment of the intelligent wheelchair local path planning and dynamic obstacle avoidance method based on a scrolling window provided by the present invention, in step S3, the line segment... This indicates the maximum width of the obstacle within the current field of view. Point and The coordinates of a point are determined relative to the camera coordinate system; Set the image acquisition interval h, acquire the obstacle's position 1 and position 2, and obtain... , , , ; Choose any point as a reference, and combine it with the interval time. This allows us to determine the direction of motion and relative speed of the obstacle.

[0009] As a preferred embodiment of the intelligent wheelchair local path planning and dynamic obstacle avoidance method based on a scrolling window provided by the present invention, the motion direction of the obstacle is calculated, and the steps are as follows: Select the obstacle location point and set it as... ; like This indicates that the obstacle is moving to the left relative to the wheelchair; like If , it means that the direction of movement of the obstacle is to the right relative to the wheelchair.

[0010] The method will be applied to mobile devices, including at least smart wheelchairs, specifically for fully autonomous movement in complex and changing environments.

[0011] It is clear without a doubt that the technical solution described above in this application can solve the technical problem that this application aims to address.

[0012] Meanwhile, through the above technical solutions, the present invention has at least the following beneficial effects: 1. The intelligent wheelchair local path planning and dynamic obstacle avoidance method based on scrolling window provided by the present invention creates an instantaneous local map of the room based on binocular vision, plans the local path based on scrolling window to realize fully autonomous trajectory planning of intelligent wheelchair, and designs a dynamic obstacle avoidance method based on infrared / binocular vision sensor while considering various environmental constraints to improve the obstacle avoidance effect of the machine in dynamic environment.

[0013] 2. This invention is based on binocular vision to reconstruct a 3D map of an unknown and complex indoor environment and locate the intelligent wheelchair in real time. Based on this, the motion trajectory of the wheelchair from the starting point to the target point is planned. Simultaneously, a local map is created in real time using the local window method, and corresponding virtual target points are selected. The optimal path is then planned within the known local map.

[0014] 3. This invention addresses dynamic obstacles that may appear during movement by designing a binocular vision-based obstacle avoidance strategy. It identifies the direction of movement and relative speed of dynamic obstacles, transmits the rotational speed to the corresponding rotating wheel through a differential, and combines it with a servo motor to achieve circular obstacle avoidance, while fully considering all possible situations. Attached Figure Description

[0015] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 A schematic diagram showing the selection of virtual coordinate points in this invention; Figure 2 This is the local optimal path planning implementation framework of the present invention; Figure 3 This is a schematic diagram of the wheelchair arc obstacle avoidance method of the present invention; Figure 4 This is a schematic diagram of the wheelchair retraction obstacle avoidance method of the present invention; Figure 5 This is a schematic diagram of the reverse obstacle avoidance based on an infrared sensor according to the present invention; Figure 6 This is a schematic diagram of the binocular vision-based obstacle avoidance method of the present invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0018] To enable those skilled in the art to better understand the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0019] It should be noted that, unless otherwise specified, the embodiments and features and technical solutions in the present invention can be combined with each other.

[0020] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0021] Example 1 Reference Figures 1-3 A method for local path planning and dynamic obstacle avoidance of intelligent wheelchairs based on a scrolling window.

[0022] This invention addresses the autonomous trajectory planning and dynamic obstacle avoidance problems of intelligent wheelchairs. Based on binocular vision sensors and the rolling window method, it achieves autonomous localization and trajectory planning of the intelligent wheelchair by creating instantaneous local maps and reconstructing 3D maps. To solve the intelligent obstacle avoidance problem in dynamic environments, this invention considers various possible scenarios and designs obstacle avoidance strategies. Ultimately, it achieves fully autonomous trajectory planning for the machine and improves its obstacle avoidance performance in dynamic environments, thereby enhancing the mobility and quality of life for the elderly. The steps are as follows: (1) Local optimal path based on scrolling window The implementation process of the locally optimal path based on a scrolling window mainly includes: establishing a local 3D map of the room using a binocular vision sensor; traversing the entire map to determine the locations of all exits, connecting each exit to the destination, and using the shortest distance as the evaluation criterion to determine the optimal solution, which is then used as the virtual target point on the local map; based on this, planning the optimal path under this visualization map, and repeating this process until the destination. Considering the possibility of situations where there is no exit, the solution includes a plan for the wheelchair to return to the previous target point along the original route, selecting a suboptimal solution for path planning to meet actual needs.

[0023] The process of selecting virtual target points that use the shortest distance as the evaluation criterion is as follows: Figure 1 As shown: When the wheelchair is at position 1, exit 1 and exit 2 are visible within the visualization range. Connecting position 1 to exit 1 and exit 1 to the destination respectively forms Line 1. Similarly, connecting position 1 to exit 2 and exit 2 to the destination forms Line 2. Judging from the distance between the two lines, Line 1 is better, so exit 1 is selected as the virtual target point within this visualization range. When the wheelchair moves to position 2, the same method is used: connecting position 2 to exit 3 and exit 3 to the destination forms Line 3. Connecting position 2 to exit 4 and exit 4 to the destination forms Line 4. Judging from the distance between the two lines, Line 3 is better, so exit 3 is selected as the virtual target point within this visualization range.

[0024] Framework for implementing locally optimal paths based on scrolling windows, such as Figure 2 As shown.

[0025] (2) Dynamic obstacle avoidance Figure 3 The diagram shown illustrates the proposed multi-functional wheelchair's dynamic obstacle avoidance mechanism. The wheelchair uses a front-mounted infrared / binocular vision sensor to acquire boundary and position information of obstacles. (Line segments...) This indicates the maximum width of the obstacle within the current field of view. Point and The coordinates of a point are determined relative to the camera coordinate system. Based on this, the interval for acquiring images is set. h Obtain the positions 1 and 2 of the obstacle, where: , , , Choose any point as a reference, such as: Combined with the interval time This allows us to determine the direction of motion and relative speed of the obstacle. As shown in the diagram, This means that the obstacle is moving to the left relative to the wheelchair, in which case you should select... The point serves as the boundary point for the wheelchair's circular obstacle avoidance, combining the wheelchair's position with that of the obstacle. relative velocity in the direction The differential distributes rotational speed to the left and right wheels, and in conjunction with the servo motor, it achieves circular obstacle avoidance. (See diagram) The maximum radius of motion for a wheelchair to successfully avoid obstacles.

[0026] Example 2 Reference Figures 4-6 Based on the above embodiment one, a method for preventing wheelchairs from colliding with obstacles or environmental boundaries and injuring patients, limited by road width, is disclosed, and two methods are provided as follows: Method 1: Obstacle Avoidance by Retreating Based on Infrared Sensors like Figure 4 As shown in (a), four infrared sensors are installed on each of the four wheels of the wheelchair. ,sensor The distance to the boundary wall is wheelchair size When the road width is insufficient to meet obstacle avoidance requirements, the wheelchair controls its four wheels at a relative speed. Move backward, four infrared sensors Real-time monitoring of the distance between the wheelchair and the boundary wall, when the sensor Monitoring point At that time, if the sensor The detected distance is greater than And until the sensor Detected ,sensor The detected distance is still consistently greater than This means There is space between the wheelchair and the obstacle that allows the wheelchair to temporarily avoid it; at this point, the wheelchair stops and moves along... The wheelchair moves in the negative direction into a temporary avoidance space. Once the sensor on the right detects a dynamic obstacle and it has passed, the wheelchair returns to its original trajectory and continues moving forward. The process of backward obstacle avoidance based on infrared sensors is as follows: Figure 5 As shown.

[0027] Method 2: Obstacle Avoidance by Retreating Based on Binocular Vision Sensors like Figure 4 As shown in (b), its principle is the same as the previously described circular obstacle avoidance. A rotating mechanism is set on the mechanism that installs the binocular vision sensor. When the road width cannot meet the obstacle avoidance requirements, the sensor is rotated 180°, and the wheelchair controls the four wheels to move at a relative speed. Moving backward, the visual sensors observe the boundary wall. When an exit that allows for temporary obstacle avoidance appears, the left and right wheel speeds are controlled to achieve temporary obstacle avoidance. The process of backward obstacle avoidance based on binocular vision is as follows: Figure 6 As shown.

[0028] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for local path planning and dynamic obstacle avoidance of intelligent wheelchairs based on a scrolling window, characterized in that, The steps are as follows: S1: Establishing a local 3D map of the indoor area based on a binocular vision sensor; S2: Determine the local optimal path based on the established indoor local 3D map; S3: Obtains the boundary and position information of obstacles through a front-facing infrared / binocular vision sensor, obtains the position of the obstacle relative to the wheelchair, distributes the rotation speed to the left and right wheels of the wheelchair through a differential, and achieves circular obstacle avoidance in combination with the servo motor.

2. The intelligent wheelchair local path planning and dynamic obstacle avoidance method based on a scrolling window according to claim 1, characterized in that, In step S2, based on the local 3D map of the room, the entire map is traversed to determine the location of all exits, each exit is connected to the endpoint, the shortest distance is used as the evaluation criterion to determine the optimal solution, and it is used as the virtual target point of the local map.

3. The intelligent wheelchair local path planning and dynamic obstacle avoidance method based on a scrolling window according to claim 2, characterized in that, Using the virtual target point as a base, the optimal path under the visualization map is planned and repeated until the destination is reached. If there is no exit, the wheelchair is designed to return to the previous target point along the original route, and the second-best solution is selected for path planning.

4. The intelligent wheelchair local path planning and dynamic obstacle avoidance method based on a scrolling window according to claim 2, characterized in that, The steps for selecting a virtual target point using the shortest distance as the evaluation criterion are as follows: A. When the wheelchair is in position 1, select exit 1 and exit 2 as seen within the field of vision; B. Connect position 1 to exit 1 and exit 1 to the end point respectively to form Line 1. Connect position 1 to exit 2 and exit 2 to the end point respectively to form Line 2. C. Determine the distance between the two lines in step B, confirm the optimal path, and use it as the virtual target point within this visualization range; D. When the wheelchair moves to position 2, select exit 3 and exit 4 as seen within the field of vision; E. Connecting position 2 with exit 3, and exit 3 with the end point, forms Line 3; connecting position 2 with exit 4, and exit 4 with the end point, forms Line 4. F. Determine the distance between the two lines in step E, confirm the optimal path, and use it as the virtual target point within this visualization range; G. Move the wheelchair to the new position and repeat steps AF until the destination is reached.

5. The intelligent wheelchair local path planning and dynamic obstacle avoidance method based on a scrolling window according to claim 1, characterized in that, In step S3, the line segment This indicates the maximum width of the obstacle within the current field of view. Point and The coordinates of a point are determined relative to the camera coordinate system; Set the image acquisition interval h, acquire the obstacle's position 1 and position 2, and obtain... , , , ; Choose any point as a reference, and combine it with the interval time. This allows us to determine the direction of motion and relative speed of the obstacle.

6. The intelligent wheelchair local path planning and dynamic obstacle avoidance method based on a scrolling window according to claim 5, characterized in that, The steps to calculate the direction of motion of the obstacle are as follows: Select the obstacle location point and set it as... ; like This indicates that the obstacle is moving to the left relative to the wheelchair; like If , it means that the direction of movement of the obstacle is to the right relative to the wheelchair.

7. The method described in any one of claims 1-6 is applied in a mobile device that includes at least an intelligent wheelchair, specifically for fully autonomous movement in complex and changing environments.

Images