Search

RU-2861459-C1 - HARDWARE-IN-THE-LOOP SIMULATION SYSTEM FOR AUTONOMOUS NAVIGATION OF UNMANNED AERIAL VEHICLES

RU2861459C1RU 2861459 C1RU2861459 C1RU 2861459C1RU-2861459-C1

Abstract

FIELD: navigation. SUBSTANCE: hardware-in-the-loop simulation system for autonomous navigation of unmanned aerial vehicles comprises a control and monitoring module, a data conversion and transmission module, an environment and dynamics implementation module, a current position display module, a training data storage, a group autonomous navigation module, a quality indicator assessment module, and also a group of simulated autonomous vehicles from unmanned aerial vehicles, connected in a certain manner. EFFECT: expansion of the functionality of the hardware-in-the-loop simulation system for autonomous navigation of unmanned aerial vehicles by providing an assessment of quality indicators of group autonomous navigation of unmanned aerial vehicles taking into account operating conditions. 3 cl, 3 dwg

Inventors

  • Drozd Oleg Vladimirovich

Dates

Publication Date
20260505
Application Date
20251112

Claims (13)

  1. 1. A system for semi-naturalistic modeling of autonomous navigation of unmanned aerial vehicles, including a control and monitoring module, a data conversion and transmission module connected in series, a module for implementing the external environment and dynamics, a module for displaying the current position, as well as a group of simulated autonomous vehicles, including from 1 to N unmanned aerial vehicles connected to a data conversion and transmission module, characterized in that which further comprises a training data storage, a group autonomous navigation module, the inputs of which are connected to the outputs of the training data storage and the control and monitoring module, a quality indicator evaluation module, the input of which is connected to the output of the current position display module, and the output is connected to the input of the control and monitoring module, wherein the output of the group autonomous navigation module is connected to the corresponding inputs of the unmanned aerial vehicles that are part of the group of simulated autonomous vehicles.
  2. 2. A system for semi-naturalistic modeling of autonomous navigation of unmanned aerial vehicles according to paragraph 1, characterized in that the module for assessing quality indicators ensures the calculation of the following indicators of the quality of the solution of the navigation problem:
  3. – root mean square error of positioning of an unmanned aerial vehicle as part of a group;
  4. – average absolute positioning error of an unmanned aerial vehicle within a group;
  5. – average minimum distance to obstacles during active obstacle avoidance by an unmanned aerial vehicle as part of a group;
  6. – relative positioning error of the unmanned aerial vehicle;
  7. – the average reaction time of an unmanned aerial vehicle to an obstacle, from the moment the object appears until the start of the evasive maneuver;
  8. – the number of trajectory corrections per unmanned aerial vehicle in a group for a given length of the group flight route.
  9. 3. A system for semi-naturalistic modeling of autonomous navigation of unmanned aerial vehicles according to paragraph 1, characterized in that the training data storage includes a base of reference points for constructing the flight trajectory of the unmanned aerial vehicle, wherein for each reference point the following are specified:
  10. – examples of graphical visualization of the contour of the control point, taking into account the weather conditions for the use of the unmanned aerial vehicle, the visibility of the control point and the direction of approach of the unmanned aerial vehicle to the control point;
  11. – spatial coordinates of the reference point;
  12. – current spatial coordinates of the unmanned aerial vehicle at which the reference point was fixed;
  13. – the values of the linear, angular velocities and spatial orientation angles of the unmanned aerial vehicle at the moment of fixing the reference point.

Description

The invention relates to unmanned aerial vehicles and specifically to the field of software, hardware and semi-naturalistic modeling of autonomous navigation of unmanned aerial vehicles, including in conditions of distortion or suppression of the navigation field. A method for modeling unmanned vehicles and a system for implementing it are known [US 10403165 B2, G09B 9/08, G09B 9/02, G09B 9/48, published 03.09.2019]. The presented system includes a processor connected via a data bus to a flight planning system. The flight planning system includes a flight plan element calculation service located on an embedded machine-readable medium. The input data for the flight planning system are databases of unmanned vehicle route control points and route parameters specified directly by the operator responsible for flight planning. The resulting flight plan is uploaded to the control system of the simulated unmanned vehicle. The flight planning system can be deployed on stationary and portable personal computers, as well as on computing servers with remote access. The disadvantages of this system are that it does not provide: – construction of a model of autonomous navigation of unmanned vehicles, in particular, unmanned aerial vehicles; – assessment of the quality indicators of autonomous navigation of unmanned vehicles. Methods for the development and use of unmanned aerial vehicles are known [US 12292737 B2, G05D 1/00, B64C 39/02, published 06.05.2025]. The object of development is an unmanned aerial vehicle, which includes, in particular: – a navigation system including a module for planning the movement of an unmanned aerial vehicle and a flight control system for an unmanned aerial vehicle; – a set of actuators connected to the navigation system via a flight controller and correcting the flight trajectory of the unmanned aerial vehicle; – a system for capturing images of ground landmarks, directly connected to the navigation system, and including an image capture device, a stabilization and guidance device. External control signals and data from other devices recording the position and movement of the unmanned aerial vehicle in space are also received at the inputs of the unmanned aerial vehicle's navigation system. The presented methods involve the use of a drone development platform that provides access to a software module developer console, including relevant software development tools and software libraries. Software modules can incorporate machine learning models based on artificial neural networks to support autonomous flight following ground reference points. The software module developer console also provides access to a global simulation environment for testing the functionality of software modules before deployment in the drone's runtime environment. A disadvantage of the unmanned aerial vehicle development platform, which enables the implementation of these methods for the development and application of unmanned aerial vehicles, is that it does not provide an assessment of the performance indicators of autonomous group navigation of unmanned aerial vehicles. The closest in technical essence to the claimed invention are methods and systems for conducting tests to ensure accelerated development of autonomous vehicles [US 7813888 B2, G06F 19/00, G06F 17/40, G05D 1/00, published 12.10.2010]. One of the implementations of the presented system includes the following elements: – a group of simulated autonomous vehicles, which, in particular, may include from 1 to N unmanned aerial vehicles and from 1 to N unmanned ground vehicles; – module for implementing the external environment and dynamics of autonomous vehicles; – a module for displaying the current position of elements of a group of simulated autonomous vehicles; – a group of autonomous vehicles deployed directly on the test site may include from 1 to N unmanned aerial vehicles and from 1 to N unmanned ground vehicles. This allows for an assessment of the actual spatial dynamics of autonomous vehicles under operating conditions close to real ones; – a system for determining the location of ground and air autonomous vehicles within the test site; – module for converting and transmitting data on the location of autonomous vehicles; – ground-based location control modules for autonomous vehicles within the test site. Each element of the group of ground and air autonomous vehicles located directly on the test site interacts with a separate ground-based location control module for autonomous vehicles numbered 1 to N within the test site; – a charging station for autonomous vehicles; – ground control module for charging station for autonomous vehicles; – control and monitoring module; – module for recording to external data storage; – software for managing network interactions of elements of the presented system. The information interaction of the elements of the presented testing system for the accelerated development of autonomous vehicles is ensured by t