An unmanned and autonomous navigation system for inland navigation

Abstract

The application discloses an inland shipping unmanned autonomous navigation system, comprising: a route planning module for generating a target route between the starting point longitude and latitude coordinates and the target point longitude and latitude coordinates, the target route is discretely formed into a route point sequence by using a multi-constraint adaptive discretization model, and the route point sequence comprises a plurality of route points arranged along the target route; a track tracking module for calculating whether the lateral deviation of the real-time coordinates of the ship from the target route point exceeds a set threshold; a local obstacle avoidance module for real-time sensing of obstacles appearing on the target route, and generating a local obstacle avoidance path when the distance between the obstacle and the ship is less than a set safety radius; and a cooperative control module for data sharing between a plurality of unmanned ships in a ship formation. Compared with the prior art, the application solves the problems of frequent global path re-planning and insufficient obstacle avoidance capability of the existing autonomous navigation system of the unmanned ship.

Description

Technical Field

[0001] This invention relates to the field of unmanned vessel technology, and in particular to an unmanned autonomous navigation system for inland waterway transportation. Background Technology

[0002] Unmanned vessels are often combined with new energy power sources to achieve "zero-emission, low-noise" navigation. Through technological integration and model innovation, unmanned vessels are reshaping the inland waterway shipping ecosystem. Through autonomous navigation systems, unmanned vessels can achieve precise navigation and monitor the waterway environment and equipment status in real time during navigation, providing early warnings of potential dangers and improving overall transportation reliability.

[0003] When unmanned vessels rely on existing autonomous navigation systems for inland waterway transport, the complexity of the navigation environment leads to high difficulty in path planning and high requirements for obstacle avoidance capabilities. During the navigation process, unmanned vessels are prone to deviating from the planned route and are unable to avoid obstacles in a timely and smooth manner. Summary of the Invention

[0004] The purpose of this invention is to provide an unmanned autonomous navigation system for inland waterway transportation, so as to solve the problems of frequent global path replanning and insufficient obstacle avoidance capabilities of existing autonomous navigation systems for unmanned vessels.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: an unmanned autonomous navigation system for inland waterway transportation, comprising: The route planning module is used to generate the target route between the latitude and longitude coordinates of the starting point and the target point. The target route is discretized into a sequence of waypoints using a multi-constraint adaptive discretization model. The sequence of waypoints includes several waypoints arranged along the target route. The track tracking module calculates whether the lateral deviation from the target waypoint exceeds a set threshold based on the ship's real-time coordinates. If it does, it adjusts the output rudder angle to correct the course deviation. The local obstacle avoidance module is used to detect obstacles appearing on the target's flight path in real time and generate a local obstacle avoidance path when the distance to the obstacle is less than a set safety radius. The collaborative control module is used for data sharing among several unmanned vessels in a ship formation.

[0006] As a further description of the above technical solution: The route planning module includes a global route generation unit and a dynamic route adjustment unit. The global route generation unit generates an initial route between the starting point and the target point based on electronic navigation charts and real-time hydrological data through a route search algorithm. The dynamic route adjustment unit performs local route replanning on the initial route based on AIS vessel dynamic data and meteorological data through an RHC rolling optimization mechanism to obtain the target route.

[0007] As a further description of the above technical solution: The multi-constraint adaptive discretization model includes curvature-driven encryption criteria, key node reinforcement rules, and hydrological response scaling mechanism. The curvature-driven encryption criteria are used to dynamically adjust the waypoint spacing according to the curvature of the target route. The key node reinforcement rules are used to dynamically adjust the waypoint spacing in a set complex water area. The hydrological response scaling mechanism is used to dynamically adjust the waypoint spacing according to the real-time flow velocity.

[0008] As a further description of the above technical solution: When the tracking module identifies that the distance between the unmanned vessel's position and the current target waypoint is less than a set distance, it switches the target waypoint to the next waypoint.

[0009] As a further description of the above technical solution: The local obstacle avoidance module includes an environmental perception unit and an obstacle avoidance planning unit. The environmental perception unit is used to acquire multi-source sensor data to identify the position coordinates and shape dimensions of obstacles. The obstacle avoidance planning unit is used to generate temporary virtual waypoints on the outside of obstacles and insert the temporary virtual waypoints into the waypoint sequence so that the track tracking module can track the virtual waypoints.

[0010] As a further description of the above technical solution: Multi-source sensor data includes radar data, sonar data, and visual data. The multi-source sensor data is fused through a deep learning fusion model to generate a high-confidence environment map.

[0011] As a further description of the above technical solution: The collaborative control module includes a collaborative request unit and a state switching unit. The collaborative request unit generates a collaborative request including the ship ID based on the monitoring of the ship's real-time operating status and broadcasts it to establish a communication link with the nearest unmanned ship in the ship formation. After the ship receives the shared data through the laser communication link, the state switching unit adjusts the working mode of the route planning module and the local obstacle avoidance module.

[0012] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are: 1. In this invention, the route planning module of the unmanned autonomous navigation system for inland waterway transportation automatically plans the global path between the starting point and the target point and performs dynamic correction. During the navigation of the unmanned vessel, the track tracking module automatically tracks several waypoints of the global path, effectively improving the track tracking accuracy of the vessel in the complex inland waterway environment.

[0013] 2. In this invention, the unmanned vessel performs high-precision environmental perception in real time during navigation, identifies the position coordinates and dimensions of obstacles to make obstacle avoidance decisions, and avoids obstacles accurately and safely. By inserting temporary virtual waypoints near obstacles, the vessel is guided to detour in conjunction with the track tracking module. After obstacle avoidance, the vessel automatically switches back to the original waypoint sequence. By generating a local obstacle avoidance path instead of global replanning, the accuracy of waypoint tracking is ensured while minimizing the global path replanning triggered by obstacle avoidance.

[0014] 3. In this invention, a multi-constraint adaptive discretization model is used to make the waypoint density in the waypoint sequence adapt to the physical characteristics of the waterway and the dynamic environment. This avoids frequent path replanning caused by excessively small waypoint spacing, while also avoiding path deviation caused by sudden obstacles due to excessively large waypoint spacing. This effectively balances the path correction frequency and track deviation.

[0015] 4. In this invention, under unforeseen circumstances such as insufficient endurance, damage to sensing equipment, or failure of the route planning module, the collaborative control module enables unmanned vessels to utilize other unmanned vessels in the vessel formation to achieve collaborative sensing and takeover, effectively reducing the failure rate of the vessel formation. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a system architecture diagram of an unmanned autonomous navigation system for inland waterway transportation. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0019] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention. Example 1

[0020] Please see Figure 1 This invention provides a technical solution: an unmanned autonomous navigation system for inland waterway transportation, comprising: The route planning module is used to generate the target route between the latitude and longitude coordinates of the starting point and the latitude and longitude coordinates of the target point. The target route is discretized into a sequence of waypoints using a multi-constraint adaptive discretization model. The sequence of waypoints includes several waypoints arranged along the target route. The unmanned vessel sails to the target point after passing through several waypoints in the sequence of waypoints in sequence. The track tracking module calculates whether the lateral deviation (Cross-TrackError) between the ship and the target waypoint exceeds a set threshold based on the ship's real-time coordinates. If it does, it adjusts the output rudder angle to correct the course deviation. When the track tracking module identifies that the distance between the unmanned ship's position and the current target waypoint is less than a set distance, it switches the target waypoint to the next waypoint to avoid insufficient time or distance for effective correction and to avoid frequent correction. The local obstacle avoidance module is used to detect obstacles appearing on the target's flight path in real time and generate a local obstacle avoidance path when the distance to the obstacle is less than a set safety radius. The collaborative control module is used for data sharing among several unmanned vessels in a ship formation.

[0021] The route planning module includes a global route generation unit and a dynamic route adjustment unit. The global route generation unit generates an initial route between the starting point and the target point based on the electronic navigation chart (ENC) and real-time hydrological data through a route search algorithm. The dynamic route adjustment unit performs local route replanning on the initial route based on AIS vessel dynamic data and meteorological data through the RHC rolling optimization mechanism to obtain the target route.

[0022] The global path generation unit converts ENC vector data, including channel boundaries, bridge locations, fixed obstacles, navigation marks, and port facilities, into a rasterized navigation map to construct a navigable space. This is then combined with real-time hydrological data to obtain information such as water depth, current velocity vectors, and flow direction at key channel points. Based on water depth constraints and the impact of current velocity, the unit dynamically adjusts passage costs. Real-time hydrological data is acquired through hydrological sensors and meteorological stations, enabling real-time hydrological data fusion. Real-time hydrological data includes current velocity and water depth.

[0023] When the route planning module generates the target route, the Rolling Time Control (RHC) achieves dynamic path adjustment through multi-source data fusion. It receives AIS vessel dynamic data such as position, speed, heading, and vessel size of other vessels within a 5-nautical-mile radius in real time to achieve avoidance. At the same time, it integrates meteorological data such as wind speed, wind direction, and visibility to update motion constraints. Finally, the RHC rolling optimization mechanism uses a sliding time window (3-5 minutes) to perform local path replanning.

[0024] The multi-constraint adaptive discretization model includes curvature-driven encryption criteria, key node reinforcement rules, and hydrological response scaling mechanism. The curvature-driven encryption criteria are used to dynamically adjust the waypoint spacing according to the curvature of the target route. The key node reinforcement rules are used to dynamically adjust the waypoint spacing in a set complex water area. The hydrological response scaling mechanism is used to dynamically adjust the waypoint spacing according to the real-time flow velocity.

[0025] The curvature-driven encryption criterion is based on whether the radius of curvature Rc of the target route exceeds 500 m. For curved sections with Rc ≤ 500 m, the spacing between waypoints is 50-100 m to ensure high-precision tracking of vessels at sharp bends and avoid centrifugal drift. For straight sections with Rc > 500 m, the spacing between waypoints is extended to 200-300 m to reduce computational load. Forced interpolation is applied to complex waterways such as bridges, locks, and confluence areas defined in the electronic navigation chart, ensuring that the spacing between waypoints is ≤ 50 m. The hydrological response scaling mechanism dynamically adjusts the spacing between waypoints based on real-time flow velocity. Different adjustment torque coefficients K are assigned for upstream, downstream, and other conditions, resulting in a dynamic point spacing Da = Db × adjustment torque coefficient K. Db is a fixed base point spacing.

[0026] By utilizing a multi-constraint adaptive discretization model, the waypoint density in the waypoint sequence is made to adapt to the physical characteristics of the waterway and the dynamic environment. This avoids frequent path replanning caused by excessively small waypoint spacing, while also avoiding significant path deviations caused by sudden obstacles due to excessively large waypoint spacing. This effectively balances the path correction frequency and track deviation.

[0027] The local obstacle avoidance module includes an environmental perception unit and an obstacle avoidance planning unit. The environmental perception unit acquires multi-source sensor data to identify the location coordinates and dimensions of obstacles. The obstacle avoidance planning unit generates temporary virtual waypoints on the outside of obstacles and inserts them into the waypoint sequence, enabling the track tracking module to follow the virtual waypoints. The multi-source sensor data includes radar data, sonar data, and visual data. The multi-source sensor data is fused using a deep learning fusion model to generate a high-confidence environmental map.

[0028] The environmental perception unit inputs data (radar point cloud, sonar signal, and image) collected by radar, sonar, and vision into a dedicated neural network. After extracting high-level feature vectors, these vectors are concatenated and input into a fusion network for joint classification / detection. During the fusion process, the weights of each sensor are dynamically adjusted based on meteorological data collected by the route planning module. For example, the radar weight is increased in foggy weather, and the visual weight is increased in clear weather. The fusion strategy is adaptively adjusted based on environmental conditions to improve recognition accuracy, ensuring that the ship can perceive dynamic obstacles in real time under complex sea conditions. This provides reliable input for autonomous navigation and significantly reduces the frequency of global path replanning.

[0029] When an obstacle appears on the global path of the local obstacle avoidance module and the distance to the obstacle is less than the set safety radius (such as 1.5 times the ship length), the module intervenes. By inserting virtual waypoints into the waypoint sequence, the local obstacle avoidance module can generate a local obstacle avoidance path based on the virtual waypoints, adjust the ship's track tracking behavior, and return to the original waypoint sequence after the obstacle avoidance is completed.

[0030] Working principle: In the unmanned autonomous navigation system for inland waterway transportation, the route planning module automatically plans the global path between the starting point and the target point and performs dynamic correction. During the navigation of the unmanned vessel, the track tracking module automatically tracks several waypoints of the global path, effectively improving the track tracking accuracy of the vessel in complex inland waterway environments.

[0031] During navigation, the unmanned vessel performs high-precision environmental perception in real time, identifies the position coordinates and dimensions of obstacles to make obstacle avoidance decisions, and avoids obstacles accurately and safely. By inserting temporary virtual waypoints near obstacles, the vessel is guided to detour in conjunction with the track tracking module. After obstacle avoidance, it automatically switches back to the original waypoint sequence. By generating local obstacle avoidance paths instead of global replanning, the accuracy of waypoint tracking is ensured while minimizing the global path replanning triggered by obstacle avoidance. Example 2

[0032] Based on the above embodiments, this embodiment further improves upon the following technical solution: the path search algorithm is either the A* algorithm or the RRT algorithm.

[0033] The A* algorithm uses heuristic search to find the optimal path from the starting point to the target point in a gridded navigation map, making it suitable for structured waterways (such as canals) and ensuring optimality. The RRT algorithm generates feasible paths through random sampling and tree expansion, making it suitable for complex waterway environments and avoiding local minima. Example 3

[0034] Based on the above embodiments, this embodiment further improves upon the following technical solutions: The collaborative control module includes a collaborative request unit and a state switching unit. The collaborative request unit generates a collaborative request including the ship ID based on the monitoring of the ship's real-time operating status and broadcasts it to establish a communication link with the nearest unmanned ship in the ship formation. After the ship receives the shared data through the laser communication link, the state switching unit adjusts the working mode of the route planning module and the local obstacle avoidance module.

[0035] After receiving the broadcast of the cooperation request, other unmanned vessels in the convoy send shared data to the requesting vessel. The other unmanned vessels in the convoy then report their own position coordinates. The requesting vessel (follower vessel) then selects the nearest other vessel (leader vessel) to establish a laser communication link. The lead vessel then sends shared data to the follower vessel. The shared data includes environmental data (obstacle locations, water depth, current velocity, etc.) and waypoint sequences. Sharing environmental data builds collaborative perception, and sharing waypoint sequences reduces the power consumption of the follower vessel's path planning. Therefore, the follower vessel can use a state switching unit to control the path planning module to be inactive and the local obstacle avoidance module to operate at low power (only the Ka-band radar remains operational).

[0036] In the event of unforeseen circumstances such as insufficient endurance, damage to sensing equipment, or failure of the route planning module, the collaborative control module enables unmanned vessels to utilize other unmanned vessels in the fleet to achieve collaborative sensing and takeover, effectively reducing the failure and disengagement rate of the fleet.

[0037] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. An unmanned autonomous navigation system for inland waterway transportation, characterized in that, include: The route planning module is used to generate the target route between the latitude and longitude coordinates of the starting point and the target point. The target route is discretized into a sequence of waypoints using a multi-constraint adaptive discretization model. The sequence of waypoints includes several waypoints arranged along the target route. The track tracking module calculates whether the lateral deviation from the target waypoint exceeds a set threshold based on the ship's real-time coordinates. If it does, it adjusts the output rudder angle to correct the course deviation. The local obstacle avoidance module is used to detect obstacles appearing on the target's flight path in real time and generate a local obstacle avoidance path when the distance to the obstacle is less than a set safety radius. The collaborative control module is used for data sharing among several unmanned vessels in a ship formation.

2. The unmanned autonomous navigation system for inland waterway transportation according to claim 1, characterized in that, The route planning module includes a global route generation unit and a dynamic route adjustment unit. The global route generation unit generates an initial route between the starting point and the target point based on electronic navigation charts and real-time hydrological data through a route search algorithm. The dynamic route adjustment unit performs local route replanning on the initial route based on AIS vessel dynamic data and meteorological data through an RHC rolling optimization mechanism to obtain the target route.

3. The unmanned autonomous navigation system for inland waterway transportation according to claim 2, characterized in that, The multi-constraint adaptive discretization model includes a curvature-driven encryption criterion, a key node reinforcement rule, and a hydrological response scaling mechanism. The curvature-driven encryption criterion is used to dynamically adjust the waypoint spacing according to the curvature of the target route. The key node reinforcement rule is used to dynamically adjust the waypoint spacing in a set complex water area. The hydrological response scaling mechanism is used to dynamically adjust the waypoint spacing according to the real-time flow velocity.

4. The unmanned autonomous navigation system for inland waterway transportation according to claim 2, characterized in that, The path search algorithm is either the A* algorithm or the RRT algorithm.

5. The unmanned autonomous navigation system for inland waterway transportation according to claim 1, characterized in that, When the tracking module identifies that the distance between the unmanned vessel's position and the current target waypoint is less than a set distance, it switches the target waypoint to the next waypoint.

6. The unmanned autonomous navigation system for inland waterway transportation according to claim 1, characterized in that, The local obstacle avoidance module includes an environmental perception unit and an obstacle avoidance planning unit. The environmental perception unit is used to acquire multi-source sensor data to identify the position coordinates and shape dimensions of obstacles. The obstacle avoidance planning unit is used to generate temporary virtual waypoints on the outside of obstacles and insert the temporary virtual waypoints into the waypoint sequence so that the track tracking module can track the virtual waypoints.

7. The unmanned autonomous navigation system for inland waterway transportation according to claim 6, characterized in that, The multi-source sensor data includes radar data, sonar data, and visual data. The multi-source sensor data is fused using a deep learning fusion model to generate a high-confidence environment map.

8. The unmanned autonomous navigation system for inland waterway transportation according to claim 1, characterized in that, The collaborative control module includes a collaborative request unit and a state switching unit. The collaborative request unit generates a collaborative request including the ship ID based on the monitoring of the ship's real-time operating status and broadcasts it to establish a communication link with the nearest unmanned ship in the ship formation. After the ship receives the shared data through the laser communication link, the state switching unit adjusts the working mode of the route planning module and the local obstacle avoidance module.

Images