22 results about "Terminal point" patented technology

The application provides a navigation method and device of a mobile robot, electronic equipment and a storage medium, and relates to the technical field of robots. The method comprises the following steps: acquiring a positioning score of the mobile robot; if it is determined that the mobile robot loses positioning according to the positioning score, rotating the head of the mobile robot to construct a local map; acquiring a global map for navigation of the mobile robot, determining a navigation target point according to the global map; determining a target terminal point in the local map according to the navigation target point, and generating a local navigation path by using the target terminal point to perform local navigation. Through the fusion of head motion control and local planning navigation strategy, the robot can flexibly adjust the viewing angle, actively acquire a wider field of view and richer environmental information in a narrow space or an environment with unstable positioning signals, improve the navigation accuracy and robustness of the robot in a complex and variable scene, and effectively cope with challenges such as positioning drift.

Embodiments of the present application provide an automatic driving path planning method fusing large model reasoning and related equipment. The method comprises: inputting scene topological description information of a target vehicle into a large language model to obtain a driving decision instruction sequence; determining a current path node and a plurality of candidate path nodes of the target vehicle, calculating a spatial passing cost and a semantic cost for each candidate path node; selecting an intermediate path node adjacent to the current path node based on the spatial passing cost and the semantic cost of each candidate path node, and selecting a next intermediate path node based on the current intermediate path node; when it is detected that the next intermediate path node is not a target terminal point of the target vehicle, repeatedly selecting the next intermediate path node until the next intermediate path node is the target terminal point; and constructing a target planning path based on the current path node and the plurality of intermediate path nodes. In this way, the safety and robustness of the vehicle driving in a complex scene can be improved.

Provided is a work machine operated automatically with reduced discomfort. The work machine includes a controller which sets target trajectory information including target points on a target path of a control target part and an inter-target-point time and makes an upper turning body and a work device perform a target motion of moving the control target part in accordance with the target path information. When judging work including the target motion to reach a predetermined stage, the controller shifts at least one of the start point and the terminal point of the next target motion and corrects the inter-target-point time based on the change from a previous movement amount to a new movement amount. The previous movement amount is the movement amount of the control target part from the start point to the terminal point before the shift, and the new movement amount is that after the shift.

The application discloses a kind of inference method, system and equipment of surrounding vehicle driving behavior intention, the inference method includes the following steps: establishing Frenet coordinate system, the driving path of target vehicle is modeled as cubic spline curve;Obtain the position (s t ,q t ) and initial heading angle θ t Of target vehicle in coordinate system, determine initial path formula (1);Obtain the calculation formula (3) of the terminal point s f ;Initial lane-changing time t prev Is preset, according to initial lane-changing time t prev And the calculation formula of terminal point s f , obtain the preset terminal point s f Of path;Obtain the calculation formula (4) of coefficient a, b, c, d, obtain current period k, according to formula (1), formula (3) and formula (4), estimate current period k adaptive lane-changing time t prev,k , according to formula (3) and t prev,k , obtain the actual terminal point s f Of path;According to the actual terminal point s f Of path, obtain three paths, calculate the probability of each path, according to the path of highest probability, infer the driving behavior intention of target vehicle.

The utility model provides a kind of pneumatic isolated environment lift door, it is related to the technical field of isolation door, solve the problem that the transmission efficiency of movable door is low and the space occupied by driving structure is large under the condition of double-layer material conveying of existing rapid roller shutter door.The utility model includes main frame, movable door that is lifted and set in main frame, a group of lifting cylinders for driving movable door lifting symmetrically installed in the left and right sides of the front of main frame, a group of lifting linear guide rails for guiding left and right when movable door lifting symmetrically installed in the left and right sides of the front of main frame, a group of edge sealing symmetrically installed in the left and right inside of the front of main frame, a group of position sensors for positioning lifting stroke are respectively installed in the upper and lower ends of edge sealing.The utility model has the beneficial effects that: the up and down terminal point of lifting is accurately controlled, the overall structure space is simplified;Better ensure the movement track when moving;It can be used together in two sets.

The present disclosure provides a road determination method, device, equipment, storage medium and program product, and relates to the field of automatic driving. The method comprises: receiving at least two initial reference lines, the initial reference lines being used to indicate a drivable road from a vehicle to a navigation terminal point; deleting the initial reference lines meeting a deletion condition to obtain remaining reference lines; in response to the number of the remaining reference lines being at least two, determining at least two candidate reference lines from the at least two remaining reference lines based on a computing power threshold; wherein the computing power threshold is an integer greater than 1; determining a driving cost value of each candidate reference line; and determining a target reference line from the candidate reference lines according to the driving cost value, the target reference line being used to determine a real driving road of the vehicle. The method can quickly select the optimal driving road by setting the deletion condition, introducing the computing power threshold and calculating the driving cost value, and provide the user with a safer, more comfortable and efficient driving experience.

The present application relates to a kind of train operation control system and method, belong to train control technical field, including train management system, trackside equipment and on-board equipment;Wherein, train management system sends train operation path command based on pre-design plan;Trackside equipment is based on the train operation path command received and calculates the pre-permission of current train;With the tail of current train as the starting point of pre-permission, along the direction of the running path in train operation path command, search risk point in order with line area as unit, if search first risk point, then continue to search, if search second risk point, then stop searching, and second risk point is as the terminal point of pre-permission;On-board equipment is based on the pre-permission received and calculates the driving permission of current train, and based on driving permission control current train operation, can ensure the premise of safety maximization pre-permission coverage range, provide sufficient operating space for on-board equipment, effectively improve line utilization efficiency and the continuity of train operation.

The utility model discloses a chute device for embedding box conveying, including slide and wear reduction acceleration strip, the slide has import, export and the conveying surface of intercommunication import and export, wear reduction acceleration strip is set on the conveying surface along the slide length direction and wear reduction acceleration strip protrudes from the conveying surface. Compared with prior art, the chute device for embedding box conveying provided by the utility model, by setting wear reduction acceleration strip on the slide, wear reduction acceleration strip protrudes from the conveying surface, lifts the embedding box to reduce the contact area of embedding box and conveying surface, and the friction coefficient of wear reduction acceleration strip is less than the friction coefficient of conveying surface, further reduces the friction resistance when embedding box sliding, makes embedding box can smoothly slide to the slide terminal point.

The application relates to an intelligent vehicle interaction method and device, a vehicle, a storage medium and a program product. The method comprises the following steps: in response to the fact that an intelligent vehicle is in a preset driving mode, obtaining an occupant emotional state of an occupant in the intelligent vehicle; generating a vehicle planning path according to the occupant emotional state and real-time road condition data; controlling the intelligent vehicle to travel according to the vehicle planning path and playing a target media resource through a playing device in a vehicle cabin of the intelligent vehicle; in response to the fact that the intelligent vehicle reaches a path terminal point of the vehicle planning path, re-obtaining the occupant emotional state and determining an emotional comparison result between the occupant emotional state obtained this time and the occupant emotional state obtained last time; generating a new vehicle planning path according to the emotional comparison result and the real-time road condition data, and returning to the step of controlling the intelligent vehicle to travel according to the vehicle planning path. The method can improve the interaction richness between the occupant in the vehicle and the vehicle travel path.

The present disclosure provides a travel time prediction method, a travel time prediction model training method, equipment and a medium, relates to the technical field of artificial intelligence, and in particular to the technical field of intelligent traffic, deep learning and map navigation. The specific implementation scheme is: constructing a bus line map and a bus station map according to road network data, bus network data and traffic condition data; obtaining a query bus route between a starting point and a terminal point, station heat information of stations in the query bus route and current time information; and predicting a bus travel time of the query bus route by using the bus line map, the bus station map, the query bus route, the station heat information and the current time information. Through the above technical scheme, the prediction accuracy of the bus travel time can be improved.

The application provides a path planning method, device, product and vehicle, and belongs to the technical field of path planning. The method comprises the following steps: determining the relative position of a current node and a terminal point according to the poses of the current node and the terminal point; and performing a node expansion strategy of a target wheel corresponding to the relative position to determine the adjacent nodes of the current node, wherein the node expansion strategy of the target wheel is used to determine the adjacent nodes when expanding the path planning based on the independent steering of the target wheel. The application aims to improve the flexibility of path planning.

The embodiment of the application provides an AGV parameter calibration method, device, equipment, medium and product, relates to the field of engineering machinery control, and the method replaces the cumbersome process that manual data collection, deviation recording in the prior art use physical tools by automatically collecting the attitude angle difference of AGV at the preset starting point and terminal point and the deviation distance of the actual terminal point of AGV and the preset straight line path, automatically optimizes the rudder wheel angle bias value based on the attitude angle difference, and automatically optimizes the zero position bias coefficient based on the deviation distance, replaces the manual parameter adjustment process of manual dependence on empirical formula or trial-and-error method in an algorithm-driven parameter optimization mode, does not need manual repeated participation in testing and adjustment, reduces manual intervention in the calibration process, shortens the calibration period, and the parameter optimization process is accurate and controllable, avoids the uncertainty caused by human experience, solves the low AGV parameter calibration efficiency problem in the prior art, and realizes the rapid and accurate calibration of AGV parameters.

This invention relates to the field of path planning technology, and more particularly to a method and device for unmanned mining truck path planning considering endpoint uncertainty. The method includes: constructing a bird's-eye view (BEV) image; extracting BEV image features based on an attention mechanism, extracting local geometric features through a convolutional attention module, and capturing global spatial topological dependencies through a multi-head self-attention module to generate a high-dimensional environmental semantic vector; constructing a conditional diffusion model, generating a probability distribution through inverse denoising, and obtaining candidate parking poses using an accelerated sampling strategy; obtaining N candidate endpoint poses based on the candidate parking poses, the current position of the excavator, and a preset safety distance, and selecting the lowest pose as the target endpoint for path planning using a safety cost function. This invention extracts environmental features through an attention mechanism and generates a multimodal parking pose distribution using a conditional diffusion model, achieving effective modeling of endpoint uncertainty and significantly improving the safety, passability, and operational efficiency of unmanned mining truck path planning in complex mining environments.

The application discloses a power distribution terminal point-to-point joint debugging method and device based on electrical quantity characteristic differentiation, and belongs to the field of power systems. The method comprises the following steps: sending a point-to-point joint debugging work order to a joint debugging device, and performing remote debugging test on the power distribution terminal after receiving the joint debugging request sent by the joint debugging device; sending remote measurement and remote signaling point-to-point test start information to the joint debugging device after the remote debugging test is qualified, and performing differentiated comparison between the response information uploaded by the power distribution terminal after differentiated load increase and a debugging scheme, so that point-to-point verification of different phase remote measurement and different position remote signaling is realized; sending remote control switch closing and opening operation instructions to the power distribution terminal after the remote measurement and remote signaling point-to-point verification is qualified, and judging whether the remote control point-to-point is qualified according to the switch position and the remote signaling position change. The method completes efficient and rapid point-to-point joint debugging test of the power distribution automation terminal through the differentiated load increase method, is beneficial to rapid commissioning of batch power distribution automation terminals, and improves the online rate of the power distribution automation system terminal.

The application provides a building pipeline multi-point path searching method and related equipment, and the method comprises the following steps: three-dimensional modeling of a plurality of source points, a terminal point and a building environment of a pipeline is respectively performed by using a grid method; the plurality of source points of the pipeline are divided by using an improved K-means algorithm according to the layout distribution of obstacles; the divided plurality of source points are searched by using an improved A* algorithm and converged to the terminal point of the pipeline. The application has the beneficial effect that the building scene can be effectively divided, and then the automatic arrangement of the pipeline between the plurality of points is completed, and the pipeline arrangement efficiency is improved.

The utility model provides a glass bottle conveying device relates to glass bottle conveying technical field, including two conveyer belts A, the starting point of one conveyer belt A is used for connecting the material box, the terminal point of another conveyer belt A is used for butt joint material pipe, and fixed mounting has the connecting frame between two conveyer belts A, the mobile conveyer belt for butt joining two conveyer belts A is inserted in the connecting frame, after the glass bottle of misplacing is transferred to the mobile conveyer belt, the mobile conveyer belt slides to the push block along the inside of connecting frame, and the glass bottle on the mobile conveyer belt is pushed to the chute by using the push plate movement, and in the chute slide to the hopper, and through the hopper vertical downward place, then mobile conveyer belt resets, and the glass bottle of misplacing does not need to adjust the position on the mobile conveyer belt, therefore the process will not occupy the mobile conveyer belt more time, will not influence the transfer of glass bottle in the material box subsequently.

Embodiments of the present application provide a human-robot collaborative working system and method, and a storage medium, and relate to the technical field of robots. The system comprises a user device and a robot. When communicating with the robot, the user device acquires a latest explored map of the robot and fuses the latest explored map with an existing map, acquires a temporary terminal point determined based on the fused map and plans a shortest path to the temporary terminal point, and guides the user to travel according to the planned path. The first robot acquires a plurality of first return combinations, generates a first target combination containing an earliest final arrival time and a corresponding communication time, and a return event. When a trigger condition of the return event is met in the exploration process, the first robot returns to the communication point in the return event to send the latest explored environment map to the user device when communicating with the user device. Through the present scheme, personnel can be safely and quickly guided to the terminal position in a communication-limited and unknown environment, and the task response efficiency is improved.

Embodiments of the present application provide a path planning method and device, equipment and a storage medium, and relate to the technical field of robots. The method comprises: determining a candidate blocking gap that blocks a task execution device from going to a terminal point based on a current environment description; simulating the robot pushing a to-be-pushed obstacle in a pushing direction and for a pushing duration based on the quality of the to-be-pushed obstacle corresponding to the candidate blocking gap, the ground friction coefficient in the current environment description, and the multi-body contact constraint between obstacles, to obtain a post-pushing environment description corresponding to the candidate blocking gap; detecting whether the task execution device is reachable from a starting point to the terminal point according to the post-pushing environment description corresponding to the target feasible gap with the lowest total consumption; if reachable, generating a passable path; otherwise, updating the current environment description to the post-pushing environment description corresponding to the target feasible gap, and returning to the step of determining the candidate blocking gap based on the current environment description. Through the present solution, the accuracy of the planned passable path can be improved.

The application is a kind of local path planning method for underwater autonomous robot. It includes the following steps: using global path planning A* algorithm as known three-dimensional global reference path; setting space sampling parameters, establishing discrete sampling space of waypoints; finding current global path matching point from global reference path according to the positioning coordinates of current robot, finding local global terminal point; carrying out hierarchical processing on sampling space, screening the last layer of sampling space points, finding the nearest local terminal point to local global terminal point; defining multiple cost functions according to discrete sampling space and obstacle information, using analytic hierarchy process to weight and sum multiple cost function values; calculating the cost value between each layer of discrete sampling space waypoints, solving the reference waypoints in each layer of discrete sampling space with the minimum cost from current position to local terminal point based on dynamic programming algorithm. The application is simple and efficient, and is free from the influence of underwater autonomous robot perception limitation on existing path planning algorithm.

Embodiments of the present application relate to the technical field of intelligent transportation, and disclose a city green wave band optimization method, device, storage medium and program product. The method comprises the following steps: finding at least one path from a starting point to a terminal point; calculating a green light synchronization rate of each path according to a vehicle departure time, a passing time between each signal light and a prescribed green light time of each signal light; and determining an optimal path according to the green light synchronization rate of each path, thereby solving the problem of overall traffic being not smooth in the prior art.

A method for connecting a shield tunnel with a station and a section line without a vertical shaft, comprising: defining various centerlines and boundaries of the station and the tunnel; taking point A as a starting point, and taking the minimum turning radius D of the shield after expansion as an arc AB, so that the tangent of point A overlaps the centerline of the station, and the terminal point B is located on the original centerline of the shield; taking AB as the centerline to draw the inner contour of the tunnel after expansion, and adjusting the position of point B until there is no conflict with the driving boundary to determine the starting point of expansion; taking point B as a starting point, and taking the minimum turning radius d of the shield before expansion as an arc BC, requiring the terminal point C to be located on the original centerline of the shield, the tangents of points B and C to coincide with the line, and the maximum offset to be not more than 30 cm; if not, the point B is moved forward; and then taking BC as the centerline to draw the inner contour of the tunnel before expansion, and adjusting point C until there is no conflict to confirm the starting point of the line adjustment. The method realizes the smooth connection of the line before and after the shield expansion through accurate arc line adjustment, avoids the vertical shaft construction, improves the efficiency, and reduces the cost and risk.

The application discloses a kind of even super-elliptic limit ring path planning methods under structured environment, first obtain structured two-dimensional prior map, rectangular fitting is carried out to the obstacle in map and each rectangular parameter is extracted;Second, determine the current most influential obstacle to the target point, according to the set porosity, determine the rectangular circumscribed super-elliptic frequency, build rectangular obstacle envelope even super-ellipse;Then establish the local coordinate system associated with the terminal point and obstacle, determine the solving direction of limit ring and solve the even super-elliptic limit ring trajectory, solve the coordinates of the current limit ring escape point;Finally, judge whether there is an obstacle between escape point and terminal point, if there is, continue to solve even super-elliptic limit ring for the next obstacle, otherwise, solve the trajectory to reach the terminal point;The method is assisted by the advantage of even super-ellipse, while maintaining flexibility and smoothness, has higher map space utilization, shorter path planning length.