A method of predicting actions of one or more participant agents in a scene is implemented by an autonomous agent in the scene. A plurality of agent models are used to generate a set of candidate future, each candidate future providing an expected action of a participant agent. A weighting function is applied to each candidate future to indicate a correlation of the candidate future in the scene. A set of candidate future is selected for each participant agent based on the indicated correlation, wherein the plurality of agent models includes a first model representing reasonable target-oriented behavior inferred from the vehicle scene and at least one second model representing alternative behavior inferred from the vehicle scene.
一种预测场景中一个或更多个参与者代理的动作的方法由场景中的自主代理实施。多个代理模型用于生成一组候选未来,每个候选未来提供参与者代理的预期动作。将加权函数应用于每个候选未来以指示该候选未来在场景中的相关性。基于所指示的相关性为每个参与者代理选择一组候选未来,其中多个代理模型包括表示从车辆场景推断的合理目标导向行为的第一模型和表示不能够从车辆场景推断的替代行为的至少一个第二模型。
Prediction and planning of mobile robots
移动机器人的预测与规划
2023-10-20
Patent
Electronic Resource
Chinese
IPC: | B60W CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION , Gemeinsame Steuerung oder Regelung von Fahrzeug-Unteraggregaten verschiedenen Typs oder verschiedener Funktion |