7results about How to "Improve trajectory tracking performance" patented technology

The invention discloses an indoor and outdoor autonomous switching unmanned intelligent agent with a redundant propulsion system and a method, and belongs to the technical field of unmanned aerial vehicles. The indoor and outdoor autonomous switching unmanned intelligent body with the redundant propulsion system comprises an unmanned intelligent body main body, two propulsion rotors are arranged in each auxiliary propeller shell, and a main rotor is arranged in a flight opening; and a high-precision hydraulic turntable is mounted at one end of the auxiliary propeller shell. The problems that an existing multi-rotor unmanned aerial vehicle is difficult to position in an indoor GPS denial environment and poor in maneuverability in a narrow space are solved, the propelling rotors can provide lift force for maintaining the height by calculating the thrust distribution scheme of the propelling rotors, yawing, pitching and rolling moments are synthesized through differential thrust, the flight control right is taken over, and the maneuverability of the multi-rotor unmanned aerial vehicle is improved. An intelligent unmanned platform which can safely, autonomously and continuously operate at an environment boundary and a system state boundary is created, and the intelligent unmanned platform can autonomously fly in indoor and outdoor complex environments and has a fault-tolerant capability.

The invention discloses a model prediction control method and system for an unmanned aerial vehicle with uncertain load mass, and the method comprises the steps: building an unmanned aerial vehicle dynamics model containing load interference, and converting the unmanned aerial vehicle dynamics model into a universal discrete state space system model; a plurality of model prediction controllers are constructed, each model prediction controller comprises a local linear model, each model prediction controller obtains a predicted value of the observation value according to the current system state prediction based on the corresponding local linear model, and all the calculated predicted values of the observation values are superposed to obtain a predicted value of the system observation value; and each model prediction controller calculates a control quantity predicted value according to the current system state based on the corresponding local linear model, and superposes all the calculated control quantity predicted values to obtain a system control quantity predicted value. According to the method, online learning and adaptation to system dynamic changes caused by load mass changes can be achieved, uncertainty is actively dealt with, and therefore robust and accurate trajectory tracking control is achieved.

The application relates to the technical field of cross-domain multi-modal robots, in particular to a land-air robot with high ground self-adaptability and a control method thereof. The land-air robot comprises an air movement unit, a ground movement unit and an automatic driving module. The air movement unit is used for providing flight power for the land-air robot, so that the land-air robot has a flight mode. The ground movement unit is arranged at the bottom of the air movement unit, so that the land-air robot has a ground movement mode. The ground movement unit comprises a movement module and a five-link wheel-foot mechanism symmetrically arranged on the left and right sides of the land-air robot. The movement module is used for adjusting the posture of the five-link wheel-foot mechanism, so that the land-air robot realizes dynamic balance and posture righting in the ground movement mode. The automatic driving module is electrically connected with the air movement unit and the ground movement unit. The land-air robot in the application can move on land and in the air, and has good trajectory tracking performance in the land-air dual mode.

This invention provides a deep learning-based method for multi-target pedestrian tracking, relating to the field of computer vision technology. The method includes: S1 acquiring video data, extracting frames from the video data to convert them into image frames, and preprocessing the image frames; S2 using an improved YOLOv5-s model to detect pedestrian targets in the image frames, obtaining the pedestrian targets; S3 using an improved Strong SORT model to track the pedestrian targets in each frame, assigning an ID to each tracked pedestrian, and generating pedestrian tracking results. This invention can accurately locate each pedestrian in a video and track their trajectory.

The application provides a path planning cooperative control method for ship-borne heterogeneous strong coupling unmanned equipment, relates to the technical field of intelligent control and path planning, and comprises the following steps: considering the longitudinal, transverse and yaw motion of an unmanned tractor and an unmanned carrier aircraft on the deck plane of an aircraft carrier, a three-degree-of-freedom coupling motion mathematical model of the heterogeneous unmanned equipment is established; a virtual force field is constructed according to an improved artificial potential field method, a field strength adjusting operator and a heading disturbance injection term are introduced, and discrete path points are generated; the discrete path points are converted into continuous expected trajectories by using a Gaussian smoothing function; precise trajectory tracking of the unmanned equipment is realized through an adaptive neural network double-layer control architecture, the upper layer controls the unmanned carrier aircraft, the lower layer controls the unmanned tractor, and cooperative control is realized by using the coupling relationship. The application introduces a first-order filter (dynamic surface control technology), effectively suppresses the differential explosion phenomenon, reduces the algorithm complexity, and improves the engineering applicability of the controller.

The application discloses a track following dynamic correction method and system for a caterpillar electric chassis, and belongs to the field of intelligent control of agricultural robots. The method is aimed at the yellow sand substrate environment of a sunlight greenhouse. Firstly, a reference trajectory for inter-ridge operation is constructed, and a turning area is calibrated. Multi-source data are collected through an encoder, an IMU, a laser radar and a visual sensor, and a chassis pose, a deviation and a caterpillar slip state are fused and estimated. A prediction control model with time sequence embedding and dimension expansion is constructed, control weights and constraint conditions are adjusted online in combination with working condition identification. A caterpillar speed correction amount is solved through rolling optimization, and is output after being limited in amplitude. Meanwhile, disturbance estimation is updated according to an execution result, so that closed-loop control is realized. The application effectively improves the trajectory tracking precision and robustness under a soft ground, suppresses deviation accumulation during turning, and guarantees stable operation of the chassis under complex working conditions.

The invention provides a flexible joint mechanical arm tracking control method based on UDE, which is applied to the field of robot intelligence, and comprises the following steps: defining a similar gap lag nonlinear phenomenon based on a gap lag phenomenon of a robot dynamic model; based on the time derivative of the tracking error and the expected error dynamic characteristics of the filtering error, in combination with a robot dynamic model and a class gap lag nonlinear phenomenon, a dynamic equation carrying uncertain items is obtained; carrying out approximation on the uncertain item to obtain an estimated value of the uncertain item; and obtaining a control model of the controller based on uncertainty and disturbance estimation by combining the estimated value of the uncertain item with the kinetic equation carrying the uncertain item. According to the method, modeling errors, parameter uncertainty and hysteresis nonlinear characteristics are collectively called'total disturbance ', so that the common complexity of modeling interference sources one by one in a traditional method is avoided, disturbance can be compensated in real time by simply adjusting bandwidth parameters of the UDE filter, and the control precision is remarkably improved.