CN-121997613-A - Polar region ship path planning simulation platform and system based on ASVSim and vision technology

CN121997613ACN 121997613 ACN121997613 ACN 121997613ACN-121997613-A

Abstract

The invention belongs to the field of ship path planning simulation, and discloses a polar region ship path planning simulation platform and a polar region ship path planning simulation system based on ASVSim and a vision technology. The polar environment module models the polar environment by utilizing a three-dimensional rendering engine, the ship model module provides a ship dynamics model, the sensor module simulates various sensors, the control decision module provides an algorithm interaction API, and the physical simulation module integrates data to form a closed loop. The invention also provides a path planning system which comprises a sensing module, a data processing module and a path planning module. The invention can simulate the natural environment of polar ship navigation realistically, and provides abundant data sources and reliable test support for the verification of a path planning system.

Inventors

XING XIANGLEI
QIAN XIAOYANG
CAI CHENGTAO
QING YANG
LU ZHIYUAN
GONG XIAOPENG

Assignees

哈尔滨工程大学

Dates

Publication Date: 20260508
Application Date: 20260212

Claims (9)

1. The polar region ship path planning simulation platform based on ASVSim and vision technology is characterized by comprising a polar region environment module, a ship model module, a sensor module, a control decision module and a physical simulation module; The polar environment module is used for carrying out three-dimensional rendering modeling on the polar environment elements by utilizing the three-dimensional rendering engine; The ship model module is used for modeling the ship dynamics model for driving the ship to move and the appearance of the scientific investigation ship; The sensor module is used for collecting data in the polar environment module in real time; the control decision module is used for receiving the data of the sensor module and sending a control instruction given by a decision algorithm to the ship model module; And the physical simulation module is used for integrating the data of the polar environment module, the ship model module, the sensor module and the control decision module, performing physical simulation and forming a closed loop.
2. The platform of claim 1, wherein the polar environment module utilizes the illusion engine UE5 to model three-dimensional rendering of ice floes, icebergs, and seawater and simulate physical collision characteristics of ice floes and icebergs; Specifically, for polar floating ice, not only a static grid body is established, but also parameterization modeling is carried out through a physical engine interface of UE5, density and quality attributes are given to each floating ice instance, a simplified multi-convex hull collision body is adopted to approximately represent the complex outline of the iceberg so as to balance calculation cost and collision precision in path planning, for sea water modeling, fluid simulation based on Gerstner wave function is carried out, and a regional vector field is superimposed, and each vertex in wave form Is the position of (c) over time The variation of (c) follows: ; Wherein, the For the direction of the i-th wave component, In order for the amplitude to be the same, Is the number of waves to be used, As a function of the position of the object, The time period of time required for the device to be in contact with the substrate, In order to be of an angular frequency, Is the initial phase.
3. The platform of claim 1, wherein the ship model module comprises an appearance blueprint and a dynamics model; The appearance blueprint is based on a snow dragon-shaped polar scientific investigation ship, a three-dimensional model of the ship is built, and a ship dynamics model provided by ASVSim plug-in units is used as a dynamics model.
4. The platform of claim 1, wherein the sensor module comprises an RGB camera, barometer, IMU, GPS, distance sensor, and radar provided by ASVSim plug-in.
5. The platform of claim 1, wherein the control decision module comprises a decision algorithm and a control vessel API, The decision algorithm is used for calculating according to the received environmental data, and finally generating a decision instruction comprising rudder angle and horsepower of the ship; The control ship API is used for realizing real-time interaction between an external algorithm environment based on Python and a three-dimensional rendering engine environment.
6. A polar region ship path planning system based on ASVSim and vision technology, which is realized by the platform of any one of claims 1-5, and is characterized in that the system comprises a perception module, a data processing module and a path planning module; the sensing module is used for acquiring environment sensing data and ship state data by using the sensor module; the data processing module is used for processing the environment sensing data to generate an obstacle map; And the path planning module is used for calculating a control strategy of the ship according to the obstacle map and the ship state data.
7. The system of claim 6, wherein the process of using the sensor module to obtain the environmental awareness data and the vessel status data comprises: The method comprises the steps of obtaining a local RGB image by using a forward-looking RGB camera, obtaining depth information by using a radar, obtaining a global RGB image by using a top RGB camera, and obtaining the positioning and the gesture of a ship by using an IMU and a GPS.
8. The system of claim 7, wherein processing the context awareness data to generate an obstacle map comprises: inputting the partial RGB image into an image segmentation model to obtain a segmentation map so as to obtain a pixel index of an obstacle under an image coordinate system, wherein the image segmentation model is a zero sample segmentation model; For pixel index , ) Inquiring the depth information synchronously acquired to obtain a depth value Acquiring an internal reference matrix of the front-view RGB camera And an extrinsic transformation matrix Using the depth value And the reference matrix Indexing the pixels , ) Converting from the image coordinate system to the camera coordinate system to obtain a camera coordinate point Using the external parameter transformation matrix Coordinate points of the camera Converting to world coordinate system to obtain three-dimensional point Combining a plurality of the three-dimensional points To form an obstacle point cloud; and carrying out orthogonal projection on the obstacle point cloud to a preset horizontal plane to generate a grid type local obstacle map.
9. The system of claim 8, wherein calculating a control strategy for a vessel based on the obstacle map and the vessel status data comprises: And (3) adopting a dynamic window method DWA, taking the local obstacle map, the global map generated by the global RGB image and the gesture, speed and dynamics constraint of the ship as inputs, outputting rudder angle and horsepower for controlling the ship, and then transmitting the rudder angle and horsepower to a UE (user equipment) end through a ship control API (application program interface) to complete real-time closed-loop control.

Description

Polar region ship path planning simulation platform and system based on ASVSim and vision technology Technical Field The invention belongs to the field of ship path planning simulation, and particularly relates to a polar region ship path planning simulation platform and system based on ASVSim and vision technology. Background With the increasing importance of the strategic value of arctic channels, it has become a core technology to plan safe and efficient paths for ships to traverse complex ice areas. Development of advanced and reliable path planning and obstacle avoidance systems is a key to achieving autonomous navigation of a ship, and effectiveness of the systems must be ensured through sufficient testing and verification. However, the verification process of the system faces a huge bottleneck, the real ship test is carried out in a real polar environment, the cost is extremely high, the environmental risk is huge, the logistic guarantee is extremely complex, experimental conditions are difficult to reproduce due to instantaneous variation of sea ice distribution, and the contrast test and iterative optimization of different systems cannot be carried out, so that the development of the polar ship intelligent technology is severely restricted. For this reason, system testing using simulation techniques has become an industry consensus, and is the most effective way to solve the above-mentioned dilemma. However, the existing general automatic driving simulation platform (such as AirSim with open source) is mainly oriented to land vehicle scenes, and generally lacks of modeling the high fidelity of the specific environmental elements of the polar region, such as the physical characteristics of sea ice, the collision dynamics of the ship body and the sea ice, the polar region weather with a plurality of ends and the like. The distortion of the simulation environment causes great gap between the system test result and the practical application, and reliable verification support cannot be provided for the effectiveness of the system. Therefore, a special simulation platform capable of realistically simulating a polar navigation scene is urgently needed in the field, and an efficient path planning system matched with the special simulation platform is developed on the basis of the special simulation platform, so that research and application of a polar ship path planning technology are promoted in a safe and low-cost mode. Disclosure of Invention The invention provides a polar region ship path planning simulation platform and a polar region ship path planning simulation system based on ASVSim and vision technology, which can simulate the natural environment of polar region ship navigation realistically and provide abundant data sources and reliable test support for the verification of the path planning system. In order to achieve the above object, the present invention provides the following solutions: the polar region ship path planning simulation platform based on ASVSim and vision technology comprises a polar region environment module, a ship model module, a sensor module, a control decision module and a physical simulation module; The polar environment module is used for carrying out three-dimensional rendering modeling on the polar environment elements by utilizing the three-dimensional rendering engine; The ship model module is used for modeling the ship dynamics model for driving the ship to move and the appearance of the scientific investigation ship; The sensor module is used for collecting data in the polar environment module in real time; the control decision module is used for receiving the data of the sensor module and sending a control instruction given by a decision algorithm to the ship model module; And the physical simulation module is used for integrating the data of the polar environment module, the ship model module, the sensor module and the control decision module, performing physical simulation and forming a closed loop. Preferably, the polar environment module utilizes the illusion engine UE5 to perform three-dimensional rendering modeling on the floating ice, the iceberg and the seawater, and simulate physical collision characteristics of the floating ice and the iceberg; Specifically, for polar floating ice, not only a static grid body is established, but also parameterization modeling is carried out through a physical engine interface of UE5, density and quality attributes are given to each floating ice instance, a simplified multi-convex hull collision body is adopted to approximately represent the complex outline of the iceberg so as to balance calculation cost and collision precision in path planning, for sea water modeling, fluid simulation based on Gerstner wave function is carried out, and a regional vector field is superimposed, and each vertex in wave form Is the position of (c) over timeThe variation of (c) follows: ; Wherein, the For the direction of the i-th wave comp