Existing approaches to automatic assembly planning often lead to combinatorial explo- sion. When the parts composing the assembly increase in number, computer-aided planning be- comes much slower than manual planning....Existing approaches to automatic assembly planning often lead to combinatorial explo- sion. When the parts composing the assembly increase in number, computer-aided planning be- comes much slower than manual planning. Efforts to reduce the computing time by taking into ac- count various constraints and criteria to guide the search for the optimal plan requires too much input information, so as to offset the convenience of automatic assembly planning. In addition, as the planner becomes more complicated, such efforts often fail to reach the objective. This paper presents a new concep── unit , asserting that the intemal structure of an assembly is hierachical. Every disassembly operation only handles several units, no matter how many parts are involved. Furthermore, the scenario of disassembly is brought to light. It relates to only two key data──the liaison type and the assembly direction. The computational cast of this approach is roughly propor. tional to the number of parts. A planner, implementing these principlcs can generate the optimal as- sembly plans dramatically faster than the known approaches.展开更多
Decomposing complex real-world tasks into simpler subtasks and devising a subtask execution plan is critical for humans to achieve effective decision-making.However,replicating this process remains challenging for AI ...Decomposing complex real-world tasks into simpler subtasks and devising a subtask execution plan is critical for humans to achieve effective decision-making.However,replicating this process remains challenging for AI agents and naturally raises two questions:(1)How to extract discriminative knowledge representation from priors?(2)How to develop a rational plan to decompose complex problems?To address these issues,we introduce a groundbreaking framework that incorporates two main contributions.First,our multiple-encoder and individual-predictor regime goes beyond traditional architectures to extract nuanced task-specific dynamics from datasets,enriching the feature space for subtasks.Second,we innovate in planning by introducing a top-K subtask planning tree generated through an attention mechanism,which allows for dynamic adaptability and forward-looking decision-making.Our framework is empirically validated against challenging benchmarks BabyAI including multiple combinatorially rich synthetic tasks(e.g.,GoToSeq,SynthSeq,BossLevel),where it not only outperforms competitive baselines but also demonstrates superior adaptability and effectiveness incomplex task decomposition.展开更多
产品设计过程冲突是不可避免的,冲突消解可以分为两个阶段,即冲突预消解和冲突检测消解。合理地进行设计任务分解与规划,将从根源上消除可能出现的过程冲突,实现冲突预消解。基于CCM_A(Cooperating Correlative Map Base on Activity)...产品设计过程冲突是不可避免的,冲突消解可以分为两个阶段,即冲突预消解和冲突检测消解。合理地进行设计任务分解与规划,将从根源上消除可能出现的过程冲突,实现冲突预消解。基于CCM_A(Cooperating Correlative Map Base on Activity)过程建模新方法,确定冲突预消解的图示表达;根据产品协同设计特点,建立相应的任务分解规划层次模型,确定了产品协同设计任务分解与规划约束规则,确定了协同设计任务的层次分解与规划方法与步骤;以设计信息协同为基础,提出设计任务间协同度的概念,并对其进行定义和量化表达;提出基于协同度的大粒度耦合任务集撕裂规划方法,并进行实例应用。展开更多
文摘Existing approaches to automatic assembly planning often lead to combinatorial explo- sion. When the parts composing the assembly increase in number, computer-aided planning be- comes much slower than manual planning. Efforts to reduce the computing time by taking into ac- count various constraints and criteria to guide the search for the optimal plan requires too much input information, so as to offset the convenience of automatic assembly planning. In addition, as the planner becomes more complicated, such efforts often fail to reach the objective. This paper presents a new concep── unit , asserting that the intemal structure of an assembly is hierachical. Every disassembly operation only handles several units, no matter how many parts are involved. Furthermore, the scenario of disassembly is brought to light. It relates to only two key data──the liaison type and the assembly direction. The computational cast of this approach is roughly propor. tional to the number of parts. A planner, implementing these principlcs can generate the optimal as- sembly plans dramatically faster than the known approaches.
文摘Decomposing complex real-world tasks into simpler subtasks and devising a subtask execution plan is critical for humans to achieve effective decision-making.However,replicating this process remains challenging for AI agents and naturally raises two questions:(1)How to extract discriminative knowledge representation from priors?(2)How to develop a rational plan to decompose complex problems?To address these issues,we introduce a groundbreaking framework that incorporates two main contributions.First,our multiple-encoder and individual-predictor regime goes beyond traditional architectures to extract nuanced task-specific dynamics from datasets,enriching the feature space for subtasks.Second,we innovate in planning by introducing a top-K subtask planning tree generated through an attention mechanism,which allows for dynamic adaptability and forward-looking decision-making.Our framework is empirically validated against challenging benchmarks BabyAI including multiple combinatorially rich synthetic tasks(e.g.,GoToSeq,SynthSeq,BossLevel),where it not only outperforms competitive baselines but also demonstrates superior adaptability and effectiveness incomplex task decomposition.
文摘产品设计过程冲突是不可避免的,冲突消解可以分为两个阶段,即冲突预消解和冲突检测消解。合理地进行设计任务分解与规划,将从根源上消除可能出现的过程冲突,实现冲突预消解。基于CCM_A(Cooperating Correlative Map Base on Activity)过程建模新方法,确定冲突预消解的图示表达;根据产品协同设计特点,建立相应的任务分解规划层次模型,确定了产品协同设计任务分解与规划约束规则,确定了协同设计任务的层次分解与规划方法与步骤;以设计信息协同为基础,提出设计任务间协同度的概念,并对其进行定义和量化表达;提出基于协同度的大粒度耦合任务集撕裂规划方法,并进行实例应用。
