Method and device for placing infrastructure sensors along a route

Abstract

Method (10) for placing infrastructure sensors along a route, characterized by the following features: - Drones equipped with lidar (21) are distributed in a spatial arrangement over the route which allows preliminary coverage of the route by the lidar (21), - the arrangement of the drones over the route is varied while their coverage (13) is evaluated, and - if the coverage (13) meets a predetermined quality criterion, the infrastructure sensors are placed according to the arrangement of the drones.

Description

[0001] The present invention relates to a method for placing infrastructure sensors along a route. The present invention further relates to a corresponding device, a corresponding computer program, and a corresponding storage medium. State of the art

[0002] The ongoing development in automation and robotics has led to significant advances in the automotive industry, particularly regarding the testing and implementation of autonomous vehicles. A key component for the safe and efficient operation of these vehicles is the precise perception of their surroundings, which is made possible by the use of various sensor technologies. The most common sensors currently include lidar (light detection and distance measurement), camera, and radar systems. Sensors of these types form the basis for acquiring the relevant data required for navigation and interaction with the environment.

[0003] Lidar sensors play a crucial role here, as they can precisely measure the spatial environment by scanning with laser beams. The point clouds acquired in this way make it possible to map the environment with high accuracy and are therefore of central importance for the development of autonomous vehicles.

[0004] In academia, algorithms have been developed to improve the optimization of sensor coverage in virtual environments. One such optimization method, referred to below as "AutoSCOOP," is described by HERMANN, David, et al. in "AutoSCOOP: Automated Road-Side Sensor Coverage Optimization for Robotic Vehicles on Proving Grounds." In: 2022 IEEE 5th International Conference on Industrial Cyber-Physical Systems (ICPS). IEEE, 2022. pp. 1-6.

[0005] Another method based on raycasting for calculating infrastructure-side monitoring sensors is described in DE102021208616A1.

[0006] US2022182793A1 also reveals the use of networked, infrastructure-side sensors.

[0007] US2023280769A1 discloses a method for arranging sensors to measure environmental factors. Controlled by a central computer, drones deliver the sensors to predetermined locations. Another computer identifies the need to reposition the sensors, which is then carried out by the delivery drones. Disclosure of the invention

[0008] Despite advances and the use of virtual optimization methods like AutoSCOOP, which enable the efficient placement of infrastructure sensors in a simulated environment, a problem remains: the discrepancy between simulated conditions and real-world circumstances. While simulation-based optimizations can provide a theoretically ideal sensor array, they cannot fully account for the multitude of real-world influencing factors such as changing light conditions, weather influences, or unforeseen obstacles. These factors can impair sensor performance and thus jeopardize the reliability and safety of autonomous vehicles.

[0009] A related problem is the time-consuming and resource-intensive need for manual adjustments. After the initial sensor placement based on virtual optimization, tests must be conducted in the real-world environment to evaluate sensor coverage and functionality. This requires extensive manual intervention to optimize the sensor arrangement and adapt it to the actual conditions.

[0010] This is complicated by the fact that the installation of infrastructure sensors is intended to be permanent. After installation, adjusting their position is therefore only possible with considerable additional effort.

[0011] The described problem is solved by a method for placing infrastructure sensors along a route, a device, a computer program and a corresponding storage medium according to the independent claims.

[0012] This approach has the advantage of bridging the gap between virtual planning and real-world conditions. By using drones to evaluate sensor coverage, adjustments to sensor positions and orientations can be made in real time, allowing for an immediate response to changing environmental conditions. This adaptability, in turn, leads to more precise coverage and improves data accuracy without the need for time-consuming and costly manual readjustments.

[0013] The precise positioning of the drones, based on feedback from the real-world environment, enables the direct transfer of optimized positions to the stationary infrastructure sensors. This results in more effective sensor coverage and reduces the need for subsequent corrections. Furthermore, the initial drone placement based on virtual optimization allows for a quick and focused start to the optimization process in the real world.

[0014] The described method not only ensures optimized spatial arrangement of the sensors, but also takes into account the number, orientation, and position of the drones to achieve the best possible coverage. This enables comprehensive and detailed environmental mapping, which in turn forms the basis for the safe navigation of autonomous vehicles.

[0015] Further advantageous embodiments of the invention are specified in the dependent patent claims. Brief description of the drawings Fig. Figure 1 shows a conventional roadside unit with WLAN access point, lidar and camera. Fig. 2, Fig. 3 and Fig. Figure 4 illustrates, by way of example, the coverage of a specific section of the route by three different arrangements of lidar sensors. Fig. Figure 5 schematically shows a method according to the invention from the initial distribution through optimization to sufficient coverage of the route. Embodiments of the invention

[0016] Fig. Figure 1 illustrates a conventional roadside unit (20) equipped with a WLAN access point and additional sensors such as a lidar (21) and a camera (22).

[0017] In the Fig. Figures 2 to 4 visualize three exemplary arrangements of lidar sensors (21) whose point clouds cover a given section of road in highly varied ways. These figures illustrate how the spatial distribution of the sensors influences the quality of the data acquisition and thus the accuracy of the environmental perception.

[0018] Fig.Section 5 outlines the optimized procedure (10), which begins with an initial distribution (11) of the sensor units. This initial arrangement can be based on prior virtual optimizations such as AutoSCOOP to create an efficient starting point for the subsequent real-time evaluation. This optimization (12) is performed in successive steps, using the drones to assess and adjust the sensor coverage. The goal is to achieve sufficient coverage (13) where the sensors are positioned to ensure complete and precise coverage of the route.

[0019] During the optimization phase (12), the lidar point cloud acquired by the drones is analyzed to vary the sensor positions so that the coverage is continuously improved by controlling the drones according to AutoSCOOP based on the point cloud and navigating them to specific positions. This optimization step is repeated until the route is completely or optimally covered.

[0020] Once sufficient coverage (13) is achieved, the final arrangement of the roadside units is determined; these units may also be equipped with lidar, radar, or cameras, for example. In the example shown, the optimized arrangement allows one of three infrastructure sensors to be eliminated without compromising coverage. Reference symbol list 10 procedures 11 Initial distribution 12 Optimization 13 sufficient coverage 20 Roadside Unit 21 Lidar 22 Camera QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] DE 102021208616A1

[0005] US 2022182793A1

[0006] US 2023280769A1

[0007] Cited non-patent literature

[0000] 2022 IEEE 5th International Conference on Industrial Cyber-Physical Systems (ICPS). IEEE, 2022. pp. 1-6

[0004]

Claims

[1] Method (10) for placing infrastructure sensors along a route, characterized by the following characteristics: - Drones equipped with lidar (21) are distributed in a spatial arrangement over the route which allows preliminary coverage of the route by the lidar (21), - the arrangement of the drones over the route is varied while their coverage (13) is evaluated, and - if the coverage (13) meets a predetermined quality criterion, the infrastructure sensors are placed according to the arrangement of the drones. [2] Method (10) according to claim 1, characterized by the following characteristic: - the initial arrangement of the drones is calculated using AutoSCOOP (11). [3] Method (10) according to claim 1 or 2, characterized by the following characteristics: - the evaluation of the coverage (13) is carried out using a point cloud captured by the drones using the lidar (21) and - the arrangement is varied by controlling the drones according to AutoSCOOP based on the point cloud (12). [4] Method (10) according to any one of claims 1 to 3, characterized by the following characteristic: - varying the arrangement involves the number, orientation and position of the camera drones with respect to the route (10). [5] Method (10) according to any one of claims 1 to 4, characterized by at least one of the following characteristics: - the infrastructure sensors include a lidar (21), - the infrastructure sensors include a camera (22) or - The infrastructure sensors include a radar. [6] Method (10) according to any one of claims 1 to 5, characterized by the following characteristic: - the placed infrastructure sensors form a Roadside Unit (20). [7] Device, characterized by the following characteristics: - the device is set up to carry out a method (10) according to one of claims 1 to 6. [8] Computer program which is configured to perform all steps of a method (10) according to any one of claims 1 to 6. [9] Machine-readable storage medium with a computer program stored thereon according to claim 8.

Images