47 results about "Environmental modelling" patented technology

The invention discloses a planning method for a mixed path of a mobile robot under a multi-resolution barrier environment. The invention aims to solve the problems of blindness of initial planning, environment modeling lack of flexibility and poor real-time obstacle avoidance capability of the present method. According to the technical scheme, a self-adapting inhomogeneous polar-radius polar coordinate modeling method is adopted for performing environment modeling on the motion space of the mobile robot; a gravity particle swarm searching method is adopted for planning an initial overall path from starting point to ending point; according to the initial overall path, a modified artificial potential field method is adopted for performing local dynamic obstacle avoidance by estimating the minimum safe distance and safe collision-preventing angle and for arriving at each initial overall path point in turn; and a final overall collision-free path is output after arriving the planning end point. According to the planning method provided by the invention, the blindness of initial overall planning and the environment modeling flexibility can be effectively improved, the real-time obstacle avoidance capability for dynamic unknown barrier is strong, and the method is high in speed, high in precision and strong in adaptability.

The invention provides a method for planning tasks of unmanned aerial vehicles on the basis of Q(lambda) algorithms. The method includes a step of carrying out environment modeling, a step of initializing Markov decision process models, a step of carrying out Q(lambda) algorithm iterative computation and a step of computing the optimal paths according to state value functions. The method particularly includes initializing grid spaces according to the minimum flight path section lengths of the unmanned aerial vehicles, mapping coordinates of the grid spaces to obtain airway points and representing circular and polygonal threat regions; building Markov decision models, to be more specific, representing flight action spaces of the unmanned aerial vehicles, designing state transition probability and constructing reward functions; carrying out iterative computation on the basis of constructed models by the aid of the Q(lambda) algorithms; computing each optimal path of the corresponding unmanned aerial vehicle according to the ultimate convergent state value functions. The unmanned aerial vehicles can safely avoid the threat regions via the optimal paths computed according to the ultimate convergent state value functions. The method has the advantages that the traditional Q learning algorithms and effectiveness tracking are combined with one another, accordingly, the value functionconvergence speeds can be increased, the value function convergence precision can be enhanced, and the unmanned aerial vehicles can be guided to avoid the threat regions and autonomously plan paths.

The invention discloses a visual feedback platform improving the virtual reality immersion degree. The visual feedback platform is characterized by comprising a robot, a visual servo tracking control unit and an immersion type visual feedback unit and forms a closed presence sense system with users; the visual servo tracking control unit is used for tracking movement behaviors of the users and carrying out environment modeling; the immersion type visual feedback unit receives the information of the movement behaviors of the users and environment modeling and then mixes the information with limb models of the robot into a 3D virtual reality interface, and the users wear virtual reality equipment to achieve man-machine interaction. The visual feedback platform has the advantages that the immersion degree of man-machine interaction is high, the hardware investment is small, the system is simple and the integration level is high, thereby being suitable for constructing man-machine interaction systems of robots of different functions.

The method discloses a modeling method based on a top-down multi-level virtual environment. The method comprises the following steps: topological layer modeling is used for reading a 3D environment file into a 2D top view, and showing the connection relation among the areas in the virtual environment by a drawing; perception layer modeling is used for dividing objects which can be perceived by a visual human into fixed objects and moving objects, and determining the perception information according to the vision scope and the distance; path layer modeling is used for providing task objects for the virtual human, and automatically searching the path information that meets the environment by an A-star algorithm. The multi-level modeling method helps perform the top-down and rough-detailed modeling analysis on the virtual environment; and the method has convenient, virtual and efficient effect, and has higher application value in such fields as group simulation, environment modeling, safe evacuation and the like.

The invention discloses a dynamic path planning method for an ant colony algorithm in a congestion environment, and the method comprises: the step 1, carrying out the environment modeling, and obtaining an environment factor representing the congestion condition in the congestion environment; and the step 2, after replacing the distance of the road in the ant colony algorithm with the environmental factor, carrying out dynamic path planning by utilizing the dynamic change of the environmental factor. According to the invention, environmental factors for reflecting traffic jam are provided, environmental factors in a congestion environment are obtained by using an environmental modeling method, and the environment factor is used for replacing the distance of the road in the ant colony algorithm to realize the dynamic path planning of the ant colony algorithm in the congestion environment. Therefore, the factor that the dynamic change of the environment congestion condition is not considered in the existing path planning algorithm is made up, the accuracy of the optimal solution of path planning in the congestion environment is improved, and the method better conforms to the actual situation of the current urban traffic congestion.

The invention provides a traffic environment modeling method which comprises the following steps: calculating the collision probability of a vehicle and an object according to the position of the vehicle, the position of the object and a distribution factor; calculating the risk level of the object according to the type, size and speed of the object; calculating a driving environment risk field according to the collision probability of the vehicle and the object and the risk level of the object; and according to the driving environment risk field, modeling is carried out to generate a risk distribution map in the traffic environment. Therefore, the complex traffic environment is normalized, different traffic environments are quantized in a unified manner, and the motion states of the traffic elements are added into environment modeling, so that potential risks are reflected.

The invention discloses an environment modeling and path planning method based on multi-sensor fusion. The method comprises the following steps: acquiring an image and laser point cloud data in a camera field of view on an unmanned vehicle at the same time; generating a dense point cloud depth map according to the image and the laser point cloud data; taking coordinate values in the dense point cloud depth map as input, and combining a radial basis function neural network to construct an environment model in a camera view field; and determining a path planning track point sequence of the unmanned vehicle by taking the minimum height and gradient of the track point as a target and taking an environment feature vector threshold value and a single-step length as constraint conditions. The laser point cloud data and the image of the camera are fused to serve as input data of environment model construction, the position information of the obstacles can serve as input weights to be applied to environment modeling, the positioning accuracy of the field concave-convex obstacles in the environment model can be improved, and the search time is shortened.

The invention provides a data-driven path planning method for an unmanned underwater vehicle, and is mainly applied to path planning of an underwater region traversal task. The method comprises the following steps: firstly, for a task area, carrying out environment modeling by utilizing a grid method, reading priori knowledge to update a modeling map, and generating an initial global path by utilizing initial map information; during execution of the task, detecting and receiving observation data in real time, carrying out online path planning analysis, deciding a global planning mode and a local planning mode based on a planning mode decision evaluation function F, if the global planning mode and the local planning mode are switched to the local planning mode, carrying out local planning according to a heuristic rule set, and returning to an initial global planning path after traversal of a local feature dense area is completed; according to the scheme, online guidance decision makingis carried out according to underwater real-time high-dimensional space observation data, global and local planning decision making is carried out on the basis of real-time detection conditions, an adaptive planning strategy is selected through planning mode switching, the method can be suitable for more complex environments, and the time efficiency is remarkably improved.

The invention discloses an automatic sorting control and decision making system based on an M-HSTPN model, and the system employs a three-layer architecture of decision-making layer-event layer-physical layer, and the system comprises a path planning decision making subsystem based on Python, an automatic sorting processing event subsystem based on HSTPNSim, and a physical simulation environment subsystem based on V-REP. The system can achieve the modeling simulation of the five types of mixed attributes in an automatic sorting process: discrete event, continuous process, time delay characteristics, random phenomenon and decision making. The system proposes a design method and design criteria for the M-HSTPN model, and achieves the modeling construction. Through the design of the HSTPNSimsoftware and an interface for Python and V-REP software, the system can process a more complex decision making problem and a more precise physical modeling problem through the powerful scientific computing and environment modeling capabilities.

The invention discloses an electric unmanned aerial vehicle inspection route planning method and device. The method comprises three steps of environment modeling, task modeling and route planning. According to the electric unmanned aerial vehicle inspection route planning method provided by the invention, carrying out spatial three-dimensional modeling on electric power facilities such as towers,lines and sites to be inspected and surrounding environments of the electric power facilities; sorting parameters such as position coordinates and shooting angles of all task points contained in an inspection task to obtain a sub-task set; and calculating an optimal route according to an environment modeling result and a task modeling result. The method has the advantages of high safety, high planning speed, high inspection efficiency and the like.

The invention discloses a three-dimensional multilayer environment model construction method based on interval estimation, which comprises the following steps: respectively acquiring environment perception data and pose estimation, and importing the environment perception data and pose estimation data into an existing three-dimensional environment modeling system; initializing an environment model according to a set resolution; enabling the the model to divide the ground into grids; waiting for obtaining new sensing data; when new sensing data is obtained, converting the coordinates to obtain coordinate positions corresponding to the starting point and the midpoint of the laser; performing interval estimation to obtain delimited interval estimation of a laser end point position; associating the grids and creating corresponding empty volume elements; creating corresponding entity elements according to the interval association grids; applying a constraint to the inserted new voxel to obtain a three-dimensional environment modeling model; and meanwhile, sending the new voxel as new sensing data. The environment model representation method has the advantages of being low in space complexity, capable of being updated in real time, capable of judging the occupancy state of any space position and the like.

The invention is applicable to the field of environment tests, and provides a natural wind-sand scouring environment modeling technology comprising the steps of acquiring climate state wind directionand wind speed information; acquiring a dust mixing ratio simulation data based on an area climate-dust coupling model RegCM4-Dust; computing a dust concentration; computing a wind-sand scouring flux;and computing an included angle between the wind direction and a research object, namely a wind-sand scouring angle. The natural wind-sand scouring environment is built by the wind-sand scouring fluxand the wind-sand scouring angle. According to the natural wind-sand scouring environment modeling technology provided by the invention, a numerical simulation method is used for acquiring dust concentration data, so that the arrangement of observation instruments is avoided, a lot of costs such as equipment, maintenance and labor costs are saved, and the observation time is saved; through comparing with a laboratory dust test, the technology provided by the invention can be used for assessing and evaluating the adaptation and durability of material to the natural wind-sand scouring environment, can provide a simulation environment support for the design, application and protection of the wind-sand scouring resistant material, and fills in the blank of the natural wind-sand scouring environment modeling simulation technology.

The invention discloses a scene map point and image frame matching method in environment modeling. The method mainly adopts a monocular camera to research inverse depth estimation of map points, the inverse depth of the map points is expressed as a depth filter under Gaussian distribution assumption, and inverse depth observation information is utilized to update the map points; meanwhile, a key frame selection strategy and an observation-updating mechanism based on the depth filter are utilized, extraction and inverse depth observation of the map points are researched emphatically, map point matching is improved by adopting a rough-accurate matching mode, and in terms of setting of the real world and robustness, the semi-dense reconstruction scene after inverse depth optimization is more detailed, and the depth estimation is more accurate.

The invention relates to the technical field of test systems, and further discloses a test evaluation system based on the data. The guiding and adjusting end software module comprises a statistical score design. The statistical scoring design comprises a target detection sub-module, an effectiveness evaluation sub-module, a correct rate evaluation sub-module, a target position evaluation sub-module, a task score evaluation sub-module and the like. By setting statistic score setting, evaluation mainly comprises two evaluation bases, namely target detection and environment modeling; correspondingly, the system evaluates two subjects of each piece of to-be-tested software, namely target detection and environmental modeling, the examination question data of the two subjects are different and non-versatility, the automatic evaluation module is an independent module relative to the examination process; the to-be-tested software returns the answer to the server, then the evaluation module canoperate independently without interacting with the server, and only after the evaluation module calculates the score, the persistence related interface of the server is called to store the score in the database, so that the integrity of the data during transmission flow is ensured.

The invention discloses a reservoir patrol route determination method and a terminal, and the method comprises the steps: obtaining a plurality of patrol items of a to-be-patrolled reservoir, and obtaining a patrol item set; environment modeling is carried out on the patrol item set by using a grid method to obtain an environment matrix, and the environment matrix is composed of grids; the environment factors are determined based on the environment matrix, the optimal patrol route is determined by using the optimized ant colony algorithm based on the environment matrix and the environment factors, intelligent determination of the reservoir patrol route is achieved, and the environment matrix can simulate the real environment of the reservoir, including a feasible area, a dangerous area and the like. The environment factors can reflect the environment condition around the patrol item, the environment factors are introduced into the optimized ant colony algorithm to determine the optimal patrol route, multi-target optimal route determination can be achieved, the actual environment factors of the reservoir are combined, the accuracy of the optimal patrol route is improved, and therefore efficient reservoir patrol is achieved.

The invention provides a fish growth rate prediction method and system based on environment modeling. The method comprises the following steps: acquiring environment data; based on the environment data, constructing an adaptive time mode network to obtain a water temperature prediction reasoning model; and inputting experience data of to-be-predicted fish into the water temperature prediction reasoning model to obtain a prediction result of the growth rate of the to-be-predicted fish. The water temperature prediction reasoning model is constructed by collecting the comprehensive environmental data, the prediction results of the growth rates of different fishes are obtained by combining the growth experience values of the different fishes, and the method has the advantages that the application range is wide, the environmental factors are comprehensively covered, and the prediction model does not need to be repeatedly trained.

The invention discloses a method for improving reliability of monocular vision localization through environmental model integrity assessment. The method comprises, firstly, establishing an environmentexpression model; secondly, performing integrity assessment on the established environment expression model; lastly, according to the result of integrity assessment, determining whether environment modeling is completed, if so, stopping searching. Therefore, the data integrity of a finally established environment model can be improved. According to the method for improving reliability of monocular vision localization through environmental model integrity assessment, by means of the model with rich environment data, the association probability of correct data in the environment model can be improved, and further reliability of monocular vision localization can be improved.