71 results about "Collision probability" patented technology

The invention relates to the technical field of equipment control, and provides an autonomous tracking anti-interference control method and system.The confidence coefficient of a sensor is determined by combining the historical precision of the sensor and the current signal quality through a sensing recognition module, and the fusion weight is dynamically adjusted based on the confidence coefficient, so that the target tracking state can accurately reflect the actual motion characteristics of a target; the trajectory planning module takes an interference type, an interference degree and a target tracking state as input parameters cooperatively, and combines an extended state space and a reward function containing an anti-interference reward, so that an explored tracking trajectory can actively adapt to an interference scene; meanwhile, the predicted collision probability is compared with a probability threshold value, so that the global updating triggering opportunity is ensured to be accurate, and the collision risk caused by an unreasonable track is effectively avoided; and the control optimization module generates a targeted feed-forward compensation amount according to the interference type and the interference degree, and generates a control instruction after superposing the basic control amount, thereby realizing cooperation of anti-interference compensation and trajectory tracking control.

The invention provides a machine grabbing learning method fusing shape features, contact modeling and physical constraints, and the machine grabbing learning method is a robot grabbing posture learning method capable of differentiating. According to the method, a three-dimensional point cloud serves as input, a deep neural network based on sparse voxel convolution is adopted to extract multi-scale shape features, and potential grabbing points and attitude parameters thereof are predicted. The differential contact modeling is realized by constructing a local area of direction perception and extracting geometric features reflecting the contact relationship between the mechanical claw and the object. Furthermore, the collision probability of the grabbing posture is estimated based on an implicit neural network, and a differentiable collision detection module is constructed. Meanwhile, physical constraints such as friction closing, surface alignment and geometric symmetry are introduced to serve as regular terms to jointly optimize grabbing scores, collision risks and stability; according to the method, end-to-end training is supported, efficient, robust and deployable grabbing posture estimation is achieved, and the method is suitable for various automatic grabbing tasks such as industrial robots and service robots.

The invention discloses a building collision detection method based on BIM cooperative multi-agent reinforcement learning, and belongs to the technical field of building engineering. The method comprises the following steps: firstly, acquiring geometric data and pipeline distribution information from a preset building information model, and determining model block granularity through analysis to obtain a sub-model set; then, a multi-agent system is adopted to distribute monitoring responsibilities, edge geometrical shape data and physical connection point locations are extracted, potential conflict areas with overlapped boundaries are judged, and a collision probability distribution diagram is generated; exchanging real-time data by setting an information synchronization period between intelligent agents to obtain local conflict distribution details, introducing real-time change data to train the intelligent agents by adopting a reinforcement learning algorithm, and updating a monitoring priority and a cooperative action sequence; and if the conflict points which are not solved still exist, local collision detection is carried out by adopting a distributed computing framework, and finally, efficient and accurate building information model collision detection is realized by adaptively adjusting sub-model division and task allocation rules.

The invention provides a multi-sensor self-adaptive anti-collision method and device based on deep learning and a medium, and relates to the field of ship anti-collision, and the method comprises the steps: obtaining the ship position of a target ship and multi-class obstacle data; a self-attention mechanism is adopted, data fusion is carried out on the multiple pieces of obstacle data, and a fusion feature vector is obtained; processing the fusion feature vector by adopting YOLOv5, and determining a bounding box of the obstacle, a bounding box position corresponding to the bounding box and a bounding box speed; the bounding box position and the bounding box speed are processed through a convolutional neural network, and future trajectory data of the obstacle and the collision probability of the obstacle and the target ship are obtained; and determining the course and speed of the target ship based on the future trajectory data, the ship position and the collision probability. According to the invention, through fusion of multiple sensors and application of a deep learning algorithm, high-precision collision prediction and risk assessment can be realized under complex sea conditions and low visibility.

The invention relates to the technical field of automatic warehouse material, and discloses an intelligent goods allocation distribution system for automatic warehouse logistics, and the system comprises a goods allocation data collection module; a dynamic state detector; a multi-source data fusion unit; a three-dimensional storage modeling module; intelligently distributing a strategy engine; and a PLC control interface. According to the intelligent goods allocation allocation and scheduling system for automatic warehouse logistics, a goods allocation allocation scheme is generated through a multi-target Pareto optimization algorithm, the balance between carrying distance minimization min sigma alpha * Di and space utilization rate maximization Ut is achieved, and compared with a proportional static allocation rule, the warehouse space utilization rate is increased to 82%; a dynamic grid map technology is integrated, and a track cross risk area (red early warning grid) of the stacker and the AGV is marked in real time, so that the equipment collision probability is reduced by 90%; and in combination with a goods relevance knowledge graph, special goods such as medicines are clustered and stored according to temperature zones, and the energy consumption of cross-temperature-zone operation of equipment is reduced by 35%.

The invention provides a ship collision prediction method based on shipborne navigation data, and relates to the technical field of ship monitoring, and the method comprises the steps: collecting multi-source interaction data for a multi-ship interaction region; a multi-ship motion state matrix is generated through relative motion analysis; taking the multi-ship motion state matrix as input, and constructing a multi-ship interactive game model; performing ship trajectory prediction based on the multi-ship interactive game model to generate a multi-ship prediction trajectory, and correcting the multi-ship prediction trajectory based on the real-time marine environment data to generate a multi-ship future trajectory set; and according to the multi-ship future trajectory set, performing collision probability calculation, based on a collision probability calculation result, constructing an avoidance strategy game model, performing avoidance strategy game solving, and generating a multi-ship collaborative avoidance strategy. The technical problem that the accuracy of the prediction result is insufficient and the ship navigation safety is affected due to the lack of consideration of multi-ship interaction and real-time environment in the prior art is solved, and the collision prediction accuracy and the ship navigation safety are improved.

The invention relates to the technical field of aerospace, in particular to a space debris environment quantification method based on risk entropy, and the method comprises the steps: S1, discretizing an evaluation time range into a plurality of time periods, and recording the duration of each time period as delta t; any time period tj can be expressed as tj belongs to [(j-1) * delta t, j * delta t]; j < = N, where N represents the total number of time periods; simulation is carried out, and it is assumed that N + 1 possibilities exist in the in-orbit collision disintegration condition of the space object objk; specifically, if N time periods exist, there is N possibilities for the space object to collide and disintegrate; if the space object is still not disassembled at the end of simulation, judging that N + 1 possibilities exist in collision and disassembly of the space object; s2, calculating a collision probability in a time period; and S3, calculating a risk entropy according to the collision probability, and judging the influence of the space object on the space debris environment according to the risk entropy. The space debris environment quantification method based on the risk entropy can be used for quantification of uncertainty of the space debris environment.

The invention provides a mobile robot deep reinforcement learning motion planning method based on collision probability, and belongs to the field of mobile robot navigation planning. According to the method, firstly, in order to evaluate the risk of collision between the robot and an obstacle in a complex dynamic environment, a collision probability estimation function is designed based on the relative distance and speed between the robot and the obstacle and is used for marking key obstacles around the robot; secondly, considering a key obstacle and a global motion planning target, and designing a combination form state observation space; and finally, designing a reward function for guiding the robot to avoid the key obstacle aiming at the key obstacle, bringing the information of the key obstacle into a state space and designing a reward function in a corresponding form, thereby realizing effective obstacle avoidance of the high-risk obstacle and reducing the time cost of motion planning. Meanwhile, it is verified that the algorithm improves the safe and rapid motion planning performance of the mobile robot in a complex dynamic environment in a test environment.

The embodiment of the invention discloses a vehicle collision early warning method and device based on a gradient lifting algorithm. The method comprises the steps of obtaining a historical data set corresponding to vehicle driving through an Internet of Vehicles cloud platform, constructing a vehicle collision estimation model corresponding to a historical training set based on a gradient lifting algorithm, and determining a collision estimation probability corresponding to real-time driving data according to the vehicle collision estimation model, and finally, determining a warning mode corresponding to the collision estimation probability based on the target driving style, and performing warning according to the warning mode. Multi-dimensional vehicle related data are fused through the Internet of Vehicles cloud platform, and the vehicle collision prediction model constructed based on the gradient lifting algorithm is constructed for collision probability prediction, so that the final early warning result is more accurate, the real-time requirement of real-time driving is met, and the driving experience of a target user is improved.

The invention relates to the technical field of space optimization, and discloses a container loading space optimization method based on robot collaboration, and the method comprises the steps: building a multi-dimensional physical model of a package through obtaining the point cloud data and weight distribution of the package; generating a loading scheme containing a target pose, a pressure tolerance threshold and a loading sequence based on a reinforcement learning algorithm; the mechanical arm generates a motor current-position composite control instruction according to a pre-trained grabbing dynamical model, and accurate grabbing is achieved; in the loading process, the pose of the parcel is monitored in real time through a Kalman filtering algorithm, when deviation exceeds a threshold value, a local space KD tree is constructed, a collision probability gradient field is calculated, and a compensation scheme meeting stability constraints is generated; and finally, updating the quad-tree index and feeding back to the reinforcement learning algorithm for online optimization. The problem that in the robot loading process, the grabbing force control is not accurate, and the efficiency is reduced due to execution deviation accumulation is effectively solved, and efficient and accurate automatic loading is achieved.

Technologies and techniques for determining a trajectory of an assisted-operated motor vehicle. At least one object is detected in an environment of the motor vehicle and at least one uncertainty with respect to the object is determined. A future environment with the object is predicted via an electronic computing device, as a function of the detected environment and the detected object, wherein a risk value for a planned trajectory is determined on the basis of a collision probability. A most probable impact constellation and accident severity for the most probable impact constellation is determined, wherein the collision probability and the accident severity is weighted in a risk value, and wherein the trajectory is determined as a function of the determined risk value.

This invention relates to a method for determining the collision probability of a vehicle (10) with an object (20, 25). The three-dimensional positions (x, y, z) and orientations of the vehicle (10) and the object (20) are detected. The vehicle (10) is approximated by at least one first geometry, wherein the at least one first geometry comprises a portion of the vehicle. The method is performed accordingly in the case of the object (20, 25). A first fuzziness (US1) and a second fuzziness (US2) are determined with respect to the geometry for either the vehicle (10) or the object (20, 25). A Minkowski difference (D') is formed for each combination of the at least one first geometry and the at least one second geometry. A third fuzziness (US3) is normalized by a transformation (TN). The transformation (TN) is used to calculate a computational volume (BK). The collision probability of the vehicle (10) with the object (20, 25) is determined based on the spatial union (D') of the computational volume (BK).

The invention discloses a space target collision early warning method based on GPU parallel computing and a J2 perturbation item, aims to solve the problems of high computing complexity and poor real-time performance of existing CPU processing space target collision risk assessment, and provides a multi-stage screening method based on GPU parallel computing and the J2 perturbation item. The method comprises the following steps: reading a TLE and calculating near / distant points and a reference check point at a GPU (Graphics Processing Unit); combining according to C (N, 2) and screening according to epoch, height and geocentric distance; taking a check point state as an initial value in a check interval, and adopting single-precision J2-RK iterative screening to obtain events NJ2pre and tJ2pre; the events are screened by using double-precision J2-RK in front and back check sections, and minimum distances distJ2 and tJ2 are output; and finally, SGP4 / SDP4 is used for parallel actuarial calculation to obtain a minimum distance distDC, a relative speed, a collision probability and the like. The method gives consideration to real-time performance and precision, significantly reduces the O (n) calculation cost, and is suitable for large-scale in-orbit target early warning.

An adaptive extended keeping reservation threshold probability mechanism system with resource watching window and dynamic back-off for contention-based 5G / B5G sidelink and a method thereof are disclosed. An adaptive extended Sigmoid mechanism can be applied to reduce a contention collision probability for the different types of flows contending a usage resource; a watching window can be applied to guarantee certain probability and reserve resource for a high priority flow; using dynamic truncated binary exponential (TBE) back-off mechanism can stagger the flows occurring resource collision to avoid the same consecutive collision, so as to effectively reduce contention collision probability, and improve resource usage rate and the probability of successful flow transmission. Therefore, the effect of minimizing collision probability and access delay and maximizing successful probability and throughput can be achieved.

The invention discloses an anti-collision system and an anti-collision method. The method comprises the following steps: detecting a first detection area and a second detection area through a radar to generate a detection result of a target object, wherein the second detection area comprises the first detection area and is larger than the first detection area; judging whether the target object invades the first detection area or the second detection area according to the detection result; in response to judging that the target object invades the second detection area but does not invade the first detection area, calculating a first collision probability according to the first weight and the detection result; in response to judging that the target object invades the first detection area, calculating a first collision probability according to the second weight and the detection result; and judging whether to output alarm information according to the first collision probability.

The present invention discloses a method for simulating restricted migration to calculate the flow tortuosity in a porous medium, including simplifying and correcting a sample digital core based on the watershed algorithm; including the collision probability of microscopic-scale continuous-phase particles characterized by the gray value of the digital core in different regions and different migration directions; including the migration direction of the continuous-phase operator and the digital core mirror extension mapping method for different digital core resolutions; quantifying the tortuosity in different states by simulating the stable diffusion coefficient under infinite mirror mapping and combining the operator migration mode in the real situation in the above text. Based on the digital core, the tortuosity data obtained by the method of simulating random migration in the porous medium in the present invention, compared with the existing methods of ideal mathematical models such as koponen, sousa, and barrande, introduces the complex pore structure and force influence in migration. The tortuosity obtained by the new method is simulated and calculated, and the coincidence rate of the result is increased by 11% compared with the tortuosity of the traditional method.

The invention provides a path determination method, and belongs to the technical field of path planning, and the path determination method comprises the steps: obtaining at least two candidate paths; path evaluation information of each candidate path is determined, the path evaluation information comprises at least one of collision probability, smoothness and dynamics feasibility, the collision probability represents the feasibility of collision between the candidate path and an obstacle, the smoothness represents the smoothness of the candidate path, and the dynamics feasibility represents the feasibility of collision between the candidate path and the obstacle. The dynamics feasibility represents the degree that the candidate path meets the dynamics of the target object; and determining the candidate path of which the path evaluation information meets a preset condition from the at least two candidate paths as a target path. According to the embodiment of the invention, the path most suitable for the movement of the target object can be effectively determined, so that the accurate control of the movement of the target object is realized.

The application belongs to the technical field of robot control, and discloses a mobile robot dynamic obstacle avoidance planning system and method based on visual navigation, real-time motion parameters such as obstacle position, speed and acceleration are extracted based on pose semantic perception results, which are input into a social LSTM network together with obstacle semantic types and robot motion parameters, interaction behaviors such as avoidance and following among multiple obstacles are captured through a social pooling layer, the influence of robot motion on obstacle trajectory is quantified through an interaction modeling layer, and a trajectory probability distribution and a prediction confidence of 3-5 time steps are output, while a collision probability, a minimum collision time and a minimum control amount required for collision avoidance are accurately calculated; through a lightweight Transformer double-branch shared network, a pose solution and obstacle semantic segmentation share feature extraction backbones, and robot motion constraints are integrated to realize multi-task parallel output of single-frame data, and perception link delay is shortened.

The application relates to a satellite collision probability estimation method, device and readable storage medium. The satellite collision probability estimation method comprises the following steps: forward propagation of a non-Gaussian relative velocity probability density function to obtain a reference velocity histogram, so as to determine an initial parameter set of a Gaussian mixture model; every first preset time interval, a plurality of sets of distance residuals and angle residuals of a first satellite and a second satellite in a running state are updated, so as to update the parameter set of the Gaussian mixture model; every second preset time interval, the relative positions of the first satellite and the second satellite are updated, and the instantaneous collision volume of the first satellite and the second satellite is determined by using the separation axis theorem; the mean values of Gaussian components in the updated parameter set are substituted into a static spherical harmonic coefficient table for query, so as to obtain the relative position probability density of the first satellite and the second satellite; and the collision probability is obtained based on the instantaneous collision volume and the relative position probability density, so that fast and accurate estimation is realized.

The invention belongs to the technical field of automatic driving, and particularly relates to an automatic driving complex scene safety decision-making system based on digital twinning, which comprises an environment sensing module, a data processing module, a cloud decision-making module, a digital twinning rehearsal module and a vehicle judgment and execution module, the data acquisition module is used for acquiring vehicle surrounding data and vehicle state data; the data processing module is used for processing the original data of the environment sensing module in real time to generate an obstacle information list; the cloud decision module takes security as a main target; the digital twin rehearsal module is based on a high-precision vehicle dynamics model and a safety rule base; and the vehicle judgment and execution module is used for judging whether the cloud instruction is invalid. According to the automatic driving complex scene safety decision-making system based on digital twinning, the collision probability of vehicles in a complex intersection scene can be effectively reduced, and the real-time performance and safety of automatic driving decision-making are improved.

The invention discloses a multi-form robot motion control system based on AI perception decision. The multi-form robot motion control system comprises a perception fusion module, a multi-mode prediction module, a form management module, a motion control module and a control optimization module. The invention relates to the technical field of robot intelligent control, in particular to a polymorphic robot motion control system based on AI perception decision-making, according to the scheme, discrete morphologies are linearly combined into unified hybrid dynamics through micro-hybrid weights, and a multi-modal predicted safety violation probability and entropy regularization learning morphological strategy are combined to obtain a multi-modal robot motion control system. Smooth low-frequency switching and continuous balancing of energy consumption and risks are achieved, and coupling with model prediction control is achieved; the multi-modal collision probability is merged into model prediction control cost and constraint, hybrid dynamics is adopted for solving, high-frequency whole-body inverse dynamics is mapped into constrained joint torque, safety filtering is combined, priority avoidance and smooth executable control are guaranteed, energy consumption is reduced, and the task success rate and safety are improved.

The invention belongs to the technical field of piling cars. The invention provides a reach stacker navigation system and method based on a neural network model. The method comprises the following steps: extracting first dynamic and static information of obstacles on the ground and second dynamic and static information of goods in a goods shelf area from a video image of a front target area shot by a camera; performing collision probability analysis on the second dynamic and static information by using a neural network model, performing matching according to the collision probability obtained by analysis to obtain a trajectory number, and performing planning according to the first dynamic and static information and the second dynamic and static information to obtain a plurality of advancing trajectories corresponding to the trajectory number in the front target area; and when the forward moving type piling car is about to arrive at the front target area, one of the advancing tracks is screened out to serve as a target advancing track, and the forward moving type piling car is controlled to run according to the target advancing track. The track of the forward-moving type piling car is planned based on the ground obstacles and the goods on the goods shelf at the same time, and higher navigation safety is achieved.

The invention discloses an agent dynamic obstacle avoidance method based on a probability velocity obstacle method, and belongs to the technical field of agent dynamic obstacle avoidance. According to the invention, the collision probability of the intelligent agent and the obstacle is combined with the traditional speed obstacle method, so that the intelligent agent reliably avoids the obstacle in an environment with incomplete speed information or noise; an analytic integral method and a hierarchical optimization strategy are designed and utilized, direct calculation and intelligent risk tradeoff of the collision probability are realized in a complex environment, and the calculation complexity is greatly reduced. The intelligent agent can make a decision in time, and can realize a real-time obstacle avoidance function for a single obstacle or a plurality of obstacles in actual movement.

Some implementations described herein relate to a system for generating identifiers. The system receives, from a device, a request to generate a plurality of identifiers. The system determines, based on receiving the request and by performing a parameter locking operation, a compute region parameter associated and a timestamp parameter. The system generates, based on the compute region parameter and the timestamp parameter, a set of one or more identifiers of the plurality of identifiers. Respective portions of each identifier, of the set of one or more identifiers, may indicate the compute region parameter, a version parameter associated with the parameter locking operation, the timestamp parameter, a counter value parameter, and an action parameter. The system processes, using a scrambling technique, the set of one or more identifiers and provides the set of one or more identifiers to the device.

The present application discloses a method, device, equipment and medium for sorting and conveying equipment directly distributed by a power supply station. The present application obtains a barcode collision probability model for each first transmission path; the barcode collision probability model is obtained based on the historical equipment transportation data statistics of all the first transmission paths; the first equipment to be packaged is classified, and the first equipment is placed on the corresponding conveyor belt according to the classification result; the classification result includes the position to be placed of the first equipment; with the conveyor belt as the starting point and the position to be placed as the end point, the optimal transmission path for the first equipment is searched from all the first transmission paths according to the barcode collision probability model. Through the present application, the equipment transportation efficiency can be improved while reducing the probability of the barcode falling or being damaged due to collision, thereby improving the accuracy of equipment sorting.

Techniques for determining a collision probability between a candidate trajectory associated with a vehicle and an object (e.g., a vehicle or pedestrian) in the vehicle's environment. In some cases, the techniques include selecting a set of sampled states, where each sampled state represents a predicted state (e.g., a predicted position and / or orientation) of the object at a future time. The set of sampled states may then be used to determine the collision probability. In some cases, the set of sampled states associated with a future time t may be determined based on: (i) a probability distribution (e.g., a Gaussian distribution) associated with the predicted object state at the future time t, and / or (ii) a covariance of the probability distribution. In some cases, the set of sampled states selected based on a distribution includes a set of sigma points each associated with a probability.

This invention provides a collision avoidance method and system for the reclaimer arm of a stacker-reclaimer, relating to the field of data processing technology. The method includes acquiring first, second, and third information; performing three-dimensional modeling based on the first information to obtain a three-dimensional model; simulating the stacker-reclaimer's operating scenario based on the second, third, and three-dimensional model to obtain the simulation results; obtaining fourth information; and obtaining a target collision avoidance scheme. This invention, through operating scenario simulation and risk simulation processing of the stacker-reclaimer, can pre-simulate operational conditions, accurately identify potential collision risks, effectively reduce the probability of collisions between the reclaimer arm and surrounding equipment, and utilizes a deep learning model for multi-objective optimization. The resulting target collision avoidance scheme not only avoids operational interruptions caused by frequent collision avoidance or accidental collisions but also enables the stacker-reclaimer to complete reclaiming tasks more smoothly and efficiently, reducing waiting time and operational adjustments, and improving operational efficiency.

The embodiment of the invention aims at providing an unmanned aerial vehicle control method and device for stock ground three-dimensional modeling and a medium. The method comprises the steps that multiple sets of collection points matched with a target area are generated according to the attribute of the target area to be modeled; wherein the number of collection points included in each group of collection points is different; performing iterative optimization on the multiple groups of collection points by adopting a genetic algorithm, and selecting a group of collection points with the maximum fitness in an iterative optimization result as target collection points; and planning an optimal path based on the target acquisition point, and controlling the target unmanned aerial vehicle to execute a modeling task according to the optimal path. According to the method, the number and the positions of the optimal collection points are calculated by carrying out collection point analysis iteration on the target area, so that the optimal path of flight modeling of the unmanned aerial vehicle is planned, the time required by blind repeated flight when the unmanned aerial vehicle executes a task is greatly shortened, the number of required unmanned aerial vehicles is reduced, and the collision probability is reduced; and the modeling task execution efficiency is improved.

This invention relates to the field of intelligent warehousing system technology, specifically disclosing an obstacle avoidance algorithm suitable for multi-wheel differential speed robot systems in warehousing environments. Based on the magnitude of the error obtained from obstacle position information, the algorithm sets collision probabilities for each location point in the environment, significantly improving the safety of the algorithm in practical applications. By outputting the optimal execution result in position space, it effectively eliminates errors in speed conversion, significantly improving the reliability and safety of obstacle avoidance, and providing a reliable obstacle avoidance solution for warehousing and logistics robots. The algorithm constructs an external contour polygon and obtains a position error shape polygon based on changes in the robot's movement position information; it combines these polygons to generate obstacle shapes containing position errors; it sets relative velocity obstacles and generates position obstacle information based on the relationship between the mass point and the combined polygon; and it sets the collision probability of the mobile robot's location area in the warehousing scenario based on the magnitude of the position obstacle and the error information in the environment.

The invention relates to a pedestrian-vehicle intelligent early warning system and method based on a mobile terminal. The system comprises a risk vehicle screening module which collects real-time position and speed data of pedestrians and vehicles, generates a position track sequence through Kalman filtering smoothing, calculates a relative distance and a speed difference, and judges through a preset threshold value to obtain a high-risk vehicle set; the collision probability calculation module extracts high-risk vehicle image features, obtains a vehicle recognition result through convolutional neural network matching, estimates real-time motion state parameters through extended Kalman filtering in combination with pedestrian trajectory prediction data, and calculates trajectory intersection points and collision risk probability; the early warning sequence generation module extracts the highest probability vehicle real-time distance to calculate the predicted collision time, if the predicted collision time is smaller than a preset threshold value, the predicted collision time is marked as an emergency level, and pushing queues are ranked and distributed according to a dynamic priority algorithm to obtain an early warning pushing sequence. By adopting the system, the data stability and the recognition accuracy can be improved, the early warning response time is shortened, and the accident probability is reduced.