4 results about "Path integration" patented technology

This application relates to the field of robotic systems technology, and provides a robot-based adaptive planning method and apparatus for optical mirror assembly paths. After establishing a scenario model of the optical system mirror assembly, the method pre-defines a standardized assembly path and generates an initial trajectory using a Data Modeling (DMP) system, forming an initial assembly mechanism that couples the robot with the optical system's sub-mirror modules. This effectively ensures the rationality and feasibility of the initial assembly path. Through improved path integration strategies and model predictive control, internal and external collaborative optimization is achieved, enabling mirror assembly path generalization, obstacle avoidance safety, and high-precision traversal of designated positions, thus enhancing the robustness of the robotic arm. In dynamic and unstructured environments, adaptive iterative optimization can be performed, continuously outputting the optimal path for optical system mirror assembly, ensuring the safety and accuracy of the optical system mirror assembly, and providing reliable technical support for the on-orbit assembly of large-aperture optical detection systems.

The application discloses a UAV-UAV cooperative path planning method based on deep reinforcement learning, and relates to the technical field of path planning. The application realizes the cooperative decision and path optimization of cross-platform multi-agent in a complex task scene by modeling the state and cooperative path planning of the UAV and the UAV task in a unified space grid, and combining the state value evaluation mechanism in deep reinforcement learning. Through task segment splitting and the construction of the coupling relationship between state nodes, the application effectively extracts the key frame state and inter-frame cooperative information in the task execution process, embeds a reinforcement learning framework on this basis, and significantly enhances the perception ability of path decision to task logic and spatial dynamics. By calculating the frame-pair structure path matrix as a guide factor for policy learning and value updating, the application can realize efficient path integration and resource scheduling of the UAV-UAV in multiple tasks and multiple time periods.

The invention discloses a dexterous hand operation strategy learning and cross-domain migration algorithm for a rich contact task, and belongs to the field of robot control and artificial intelligence. The method comprises the following steps: firstly, constructing a generative encoder network, and compressing a high-dimensional dexterous hand state to a probability potential space as a unified strategy representation; then, in the space, human demonstration semantics are used as guidance, an implicit dynamical model, a value function and a strategy network are jointly learned, the strategy network is used for preheating an optimization controller based on model prediction path integration, and zero sample generalization of a target domain is achieved; and finally, through a progressive updating mechanism, a small amount of data of the target domain is utilized to finely adjust the encoder, the domain difference is quickly captured, and adaptive migration of the strategy is realized. According to the method, the problem that the dexterous hand migrates from simulation to reality due to dynamic domain difference in a rich contact task is effectively solved, and the method has the advantages of high zero sample generalization ability and high adaptive migration speed.

The invention belongs to the field of computer graphics, and relates to a real-time medical volume data visualization method based on path tracking, which comprises the following steps: acquiring medical volume data of a current frame, constructing a volume rendering equation of the medical volume data of the current frame, and converting the volume rendering equation into a path integral form; calculating the illumination intensity of the medical volume data of the current frame by adopting a sample multiplexing method based on the path integration form of the volume rendering equation; rendering according to the illumination intensity of the medical volume data of the current frame to obtain a rendered image; according to the method, a volume rendering equation is accurately evaluated by using a path tracking method, and an optimal path sample is selected according to a resampling importance sampling method to calculate the illumination intensity of pixels, so that an illumination intensity calculation result is converged to a true value, and a high-quality visual result is generated; according to the method, similar path samples are screened out from the path samples of the medical volume data of the neighbor pixels and the historical frames and are reused, so that the actual sampling rate is reduced, and the method can be operated with real-time efficiency.