The invention discloses an unknown environment autonomous navigation system and method based on reinforcement learning. The system comprises a training module for constructing a flight decision model,and carrying out training of the model in a simulation environment based on deep reinforcement learning, an interaction module for realizing interaction between an airborne sensor of the unmanned aerial vehicle and the environment, a planning module for planning a path, and an application module for interacting with the planning module to realize navigation of the unmanned aerial vehicle in an actual unknown environment. The method comprises the steps: establishing and training a flight decision model; acquiring real environment information data; acquiring an unmanned aerial vehicle state diagram; and taking the unmanned aerial vehicle state diagram as the input of the model, outputting the action correspondingly adopted by the unmanned aerial vehicle, turning to the step two, acting theaction in the real environment, and changing the state of the unmanned aerial vehicle until the unmanned aerial vehicle reaches the destination. With unknown environment autonomous navigation system and method, autonomous navigation of an unknown environment can be carried out on the premise of no priori knowledge; and the system has the characteristics of simple structure, strong generalization ability and the like.

Unknown environment autonomous navigation system and method based on reinforcement learning

The invention discloses an aircraft high-attack-angle roll suppression method based on a depth deterministic strategy algorithm, and the method comprises the steps: building a roll suppression problemreinforcement learning environment, designing a depth deterministic strategy intelligent agent for a control target, and enabling a target aircraft model to be installed on a rolling single-degree-of-freedom release mechanism in a wind tunnel; at each time step, generating a group of empirical data according to the data acquired by the rolling single-degree-of-freedom release mechanism and a steering engine control instruction transmitted by the intelligent agent, and adding the group of empirical data to an empirical cache region for sampling of the intelligent agent to perform control strategy self-learning; and at each time step, adjusting the parameters of the four deep networks through the empirical data obtained by sampling from the empirical data cache region based on the depth deterministic strategy gradient algorithm, and transmitting a steering engine control instruction to the reinforcement learning environment. According to the method, a depth deterministic strategy gradient method is adopted, and aircraft high-attack-angle rocking optimal controller design independent of a model is achieved.

Aircraft high-attack-angle roll suppression method based on depth deterministic strategy gradient

The invention discloses an unmanned ship formation path tracking method based on deep reinforcement learning. The method comprises the steps of unmanned ship navigation environment exploration, rewardfunction design, formation pattern keeping, random braking and path tracking. The navigation environment of an unmanned ship is explored by multiple ships at the same time and environmental information is extracted; reward function design comprises design for formation composition conditions and path tracking errors; random braking can be used to optimize a training process of deep reinforcementlearning; formation pattern keeping is realized through a formation pattern control strategy; according to path tracking, the unmanned ship is controlled to move along a preset path through a leader-follower formation control strategy, the position of the unmanned ship is continuously updated, and path tracking of all the unmanned ships in the formation is achieved. An unmanned ship path point tracking model is trained in an accelerated mode through the collaborative exploration strategy, and the unmanned ship formation path tracking method is formed by combining the unmanned ship path point tracking model with the leader-follower formation control strategy.

Unmanned ship formation path tracking method based on deep reinforcement learning

The invention discloses an unmanned aerial vehicle flight control method based on imitation learning and reinforcement learning algorithms. The method comprises the following steps: creating an unmanned aerial vehicle flight simulation environment simulator; defining a basic action set of flight; classifying the trajectory data according to the flight basic actions; for each flight action, learning mapping network parameters from a flight basic action to an original action by utilizing imitation learning; counting the minimum continuous action number of each basic action; constructing an upper-layer reinforcement learning network, and adding the minimum continuous action number as punishment p of aircraft action inconsistency; in the simulator, obtaining current observation information andawards, and selecting corresponding flight basic actions by using a pDQN algorithm; inputting the state information of the aircraft into an imitation learning neural network corresponding to the flight basic action, and outputting the original action of the simulator; inputting the obtained original action into a simulator to obtain observation and awards of the next moment; and performing training by using a pDQN algorithm until the strategy network on the upper layer converges.

Unmanned aerial vehicle flight control method based on imitation learning and reinforcement learning algorithms

The invention discloses a rapid air traffic control anti-collision method based on reinforcement learning and linear programming, and the method comprises the following steps: 1, setting the number ofairplanes inside and outside a sector, and obtaining the position information of the airplanes; 2, acquiring dynamic parameters of the aircraft, performing regularization processing, and constructingan aircraft deep reinforcement learning model by utilizing the parameters; 3, training the neural network by using a DDPG algorithm, and optimizing the output deflection angle by using linear programming; 4, taking the position information of the aircraft as the input of a neural network to obtain a deflection angle for controlling the aircraft; and step 5, controlling the aircraft to fly according to the deflected angle, judging whether the aircraft can deflect or not, if so, deflecting to the destination, and otherwise, continuing to update the position information. According to the method,the deflection angle output by the neural network is optimized through linear programming, the convergence speed is increased, and good robustness is achieved.

Rapid air traffic control anti-collision method based on reinforcement learning and linear programming

The invention discloses a fixed-wing UAV(unmanned aerial vehicle) cluster control method based on reinforcement learning. The method comprises S1) training phase: establishing a stochastic UAV dynamicmodel, an actuator deep neural network and an evaluator deep neural network, and continuously collecting historical experience of interaction between an agent and environment and storing the historical experience in an experience pool; carrying out batch sampling randomly in the experience pool, continuously updating network parameters of an actuator and an evaluator and eventually forming and saving an evaluator network model; and S2)execution phase: for each wing plane, obtaining position and attitude information thereof through a sensor, loading the evaluator network model, outputting thebest rolling action of the wing plane by the evaluator based on a current system combined state, a lead plane roll angle set value being given by an operator, and repeating the steps above until finishing the flight mission. The method has the advantages of high real-time performance and adaptability and being capable of transferring a strategy obtained by training in simulation to real environment and the like.

Fixed-wing UAV(unmanned aerial vehicle) cluster control method based on deep reinforcement learning

The invention discloses a real-time path planning method for an unmanned aerial vehicle based on deep reinforcement learning. The method comprises the steps of S1, obtaining the current environment state of the unmanned aerial vehicle from a simulation environment, calculating the threat degree of a target object defense unit to the unmanned aerial vehicle according to a situation evaluation model, and constructing a situation map of a task area of the unmanned aerial vehicle; constructing a main network and a target network of the convolutional neural network and the competitive neural network to perform action selection; S2, obtaining the current environment state of the unmanned aerial vehicle according to the communication link, calculating a threat value of the target object defense unit to the unmanned aerial vehicle according to the situation evaluation model, constructing a situation map of the task areas of the unmanned aerial vehicle, constructing a competitive dual-Q network, loading the trained network model, evaluating the Q value of each action in the current state, selecting the action corresponding to the maximum Q value, determining the flight direction of the unmanned aerial vehicle, and completing the flight task. According to the invention, the autonomous decision-making ability of the unmanned aerial vehicle can be effectively improved, and the method has high robustness and application value.

Real-time path planning method for unmanned aerial vehicle based on deep reinforcement learning

Handle everyday research tasks with reliable, citation-backed results

Your personal Research Agent to handle research tasks with citation-backed results

Popular Tasks used by Researchers

How can I help with your research?

Meet SciSpace

Get more enhanced response by uploading the PDFs you want me to reference.

No relevant PDFs in your library

SciSpace is the AI research assistant for academics. Run systematic literature reviews on 280M+ papers, and write papers with cited sources. Trusted by 1M+ students, PhDs & researchers.

SciSpace | AI for Research

Analyze PDFs

Code & Manuscripts

Funding & Grants

Literature & Patents

Medical & Clinical Data

Systematic Review

Visualize & Present

Web & Data

Build a Google Scholar-like website for your research.

Build a website

Create charts and images for your research

Create a Chart

Write a paper for submission to a journal

Draft a manuscript

Patent Search

Design eye-catching scientific posters in minutes.

Scientific Poster Generation

Systematic Literature Review

One task is running at the moment. Your messages will be shown right after.

Drag and drop or click here to browse

Loved by <highlight>1 million+</highlight> researchers

Extract a list of specific topics and their sources from unstructured text

Topics

Compare and analyze relevant papers that matches with your search

Papers

Get insights from PDFs and bookmarked papers from your library

My library

Recent searches

Try searching for:

Catch AI-generated content in scholarly and non-scholarly content

{ai} Detector

Ai Writer

Get PDF Summaries, highlighted text explanations 

Chat with PDF

Effortlessly create in-text citations and bibliographies in APA and 2,500 other formats

Citation generator

Get explanations, summaries, and answers on academic papers

Ease up your research workflow with {scispace}'s cohort of exciting AI tools

Elevate your academic writing skills and convey your ideas the way you want

Paraphraser

Explore our range of reading and writing tools

Your file is being prepared and should be ready in a few minutes. If it's a large file, it might take a bit longer. You can close this window, and we'll email you the file when it's done.

You have reached a maximum limit of <strong>{limit}</strong> columns in the table. Remove at least <strong>1</strong> column to add or create another one.

Yan Chao

Author Tools

Chat about Author

Papers

Fixed-wing UAV(unmanned aerial vehicle) cluster control method based on deep reinforcement learning

Real-time path planning method for unmanned aerial vehicle based on deep reinforcement learning