7 results about "Linear temporal logic" patented technology

This invention belongs to the interdisciplinary field of digital twins and artificial intelligence, specifically disclosing an automatic generation method for manufacturing equipment control logic based on neural-symbolic collaboration and intermediate representation verification. This method utilizes a neural-symbolic collaboration strategy, employing four-element semantic model constraints and thought chain reasoning to transform unstructured natural language process instructions into structured hierarchical state machine (HSM) intermediate representations. Through linear temporal logic (LTL) formal verification and graph theory connectivity analysis, static security verification of the control logic is performed, effectively improving the system's temporal determinism and deadlock resistance. Based on this, the system performs dynamic physical simulation in a digital twin environment, extracting spatial coordinate deviations when mechanical interference occurs, and inversely mapping these deviations into structured error correction prompts, which are then fed back to the large language model for logic reconstruction. This invention constructs a cross-modal "generation-verification-correction" closed-loop mechanism, achieving deep integration of model cognitive intent and underlying physical space constraints, ultimately deterministically compiling the verified intermediate representation into highly reliable industrial standard control code.

The application discloses a symbolic controller design method based on chino polyhedron under linear temporal logic specification, belongs to the technical field of nonlinear system control, and solves the linear temporal logic (LTL) task controller synthesis problem of a nonlinear control system. First, an accepted path is generated for a preset LTL task. Second, a chino polyhedron and a constrained chino polyhedron are used to divide a state space, the divided regions are defined as 'units', adjacent units are allowed to intersect, a unit topology graph is constructed based on the unit intersection relationship, and the realizability of the LTL accepted path is verified. Finally, if the accepted path is realizable, the global LTL task is decomposed into a plurality of local subtasks, a local symbolic model is constructed for each unit, a local symbolic controller is iteratively designed through a 'local to global' control strategy, and combination is completed, so that a symbolic controller satisfying the global LTL specification is obtained. The application supports different units to adopt a differentiated local equivalence relationship, reduces the conservativeness of a single equivalence relationship, reduces the calculation complexity of global approximation, and improves the efficiency and practicability of controller synthesis.

The application is suitable for the field of artificial intelligence and formal verification technology, and provides a conversion system and method from natural language to linear temporal logic based on self-learning, which comprises a text conversion module, an equivalence checking module, a rule induction module and a rule library management module; the text conversion module is used for generating a predicted formula according to a basic rule and a retrieval rule; the equivalence checking module is used for verifying the equivalence of the predicted formula and a target formula and extracting a failure sample; the rule induction module is used for analyzing the failure sample and automatically inducing a new conversion rule; and the rule library management module is used for storing the rules and retrieving based on semantic similarity. The system can automatically learn from conversion errors and continuously optimize the rule library. The application overcomes the dependence on artificial rules or labeled data in the prior art, realizes unsupervised adaptive learning, and significantly improves the accuracy and field adaptability of the conversion from natural language to linear temporal logic.

The application discloses a kind of robot hierarchical learning method based on linear temporal logic and parallel waypoint planning, including mapping expert demonstration trajectory into action primitive sequence by inverse dynamics model, automatically dividing subtask boundary and determining waypoint quantity benchmark, while converting primitive sequence into linear temporal logic formula and giving text logical constraint;The text semantic features of logical formula are extracted in advance using pre-training encoder;Based on the current observation state, use parallel neural network architecture to synchronously predict multiple target waypoints in the whole process, eliminate time sequence dependence and solve strategy error accumulation;Based on the current state, text semantic features and target waypoints, output hybrid action instruction to drive robot to execute tasks.The application realizes the automatic decomposition from demonstration to structured task, guides policy learning through formal logical constraint, and significantly improves the robustness and sample efficiency of long-time sequence operation through parallel waypoint planning, suitable for application in object stacking, precision assembly and other tasks.

The application provides a vehicle safety control method and device, electronic equipment and a readable storage medium, wherein the method comprises: acquiring perception data of a vehicle laser radar, and generating a safety event according to the perception data; when multiple safety events are triggered concurrently, arbitrating the priority of each safety event and the occupation relationship of each safety event to an actuator resource, and selecting a control instruction matched with the arbitration result from candidate control instructions as an executable control instruction sequence; the candidate control instruction is a linear temporal logic formula converted based on a vehicle safety specification and a formal model constructed in advance, and a control instruction satisfying the linear temporal logic formula is verified in advance; and the executable control instruction sequence is issued to an actuator for execution. In this way, by combining formal verification with an event triggering mechanism and using a priority arbitration mechanism to solve instruction conflicts, the accuracy of vehicle safety control is improved.

This invention relates to the field of robot motion planning and formalization technology, and discloses a method and system for reactive motion planning of robots. The method includes: establishing a set of atomic propositions and a shared-safe linear temporal logic task reduction based on the task to be performed by the robot; converting the task reduction into a deterministic finite automaton, constructing a convex region in the robot configuration space, and establishing a convex set state transition system; constructing a convex set graph by multiplying the convex set state transition system with the deterministic finite automaton, and solving for the initial trajectory; performing structured relaxation on the deterministic finite automaton; and reconstructing the convex set graph based on the relaxed deterministic finite automaton, starting from the current robot configuration, and solving for the replanned trajectory. This invention achieves safe, verifiable, and highly robust execution of long-term sequential tasks of robots in dynamic environments by unifying task semantic hierarchy, automaton logic relaxation, and continuous trajectory optimization of the convex set graph.

The application provides a substation relay protection device adaptive control method and system, the method comprises the following steps: constructing a mapping relationship between linear temporal logic sequences and relay protection events; detecting state change events, power grid operation events and manual operation events of each relay protection device in the substation, and assigning each event a globally unique event identifier and a timestamp to construct an unforgeable chain event log in the order of event occurrence, the chain event log stores original data and context information of all events; taking the chain event log as input, verifying through linear temporal logic sequences to correspondingly identify a target event sequence that violates the temporal constraint; analyzing the target event sequence to detect the corresponding target control action, and correspondingly generating a control path and a control object that adapt to the target control action. The application can greatly improve the control efficiency of the relay protection device.