Psychologically Inspired Planning Method for Smart Relocation Task

Abstract

Behavior planning is known to be one of the basic cognitive functions, which is essential for any cognitive architecture of any control system used in robotics. At the same time most of the widespread planning algorithms employed in those systems are developed using only approaches and models of Artificial Intelligence and don't take into account numerous results of cognitive experiments. As a result, there is a strong need for novel methods of behavior planning suitable for modern cognitive architectures aimed at robot control. One such method is presented in this work and is studied within a special class of navigation task called smart relocation task. The method is based on the hierarchical two-level model of abstraction and knowledge representation, e.g. symbolic and subsymbolic. On the symbolic level sign world model is used for knowledge representation and hierarchical planning algorithm, MAP, is utilized for planning. On the subsymbolic level the task of path planning is considered and solved as a graph search problem. Interaction between both planners is examined and inter-level interfaces and feedback loops are described. Preliminary experimental results are presented.

Application of Automated Guided Vehicles in Smart Automated Warehouse Systems: A Survey

“…Since the learning speed is low only using the ANNs, and it is impossible to learn all the states only using the Q-learning algorithm, we can combine them to overcome the disadvantages of either [133]. In general, there are five outputs for AGVs in warehouses, i.e., travel forward, stop, turn left 90°, turn right 90°and self-rotate 180° [13], Demircan et al [134] used ANNs to create route-planning controllers and then used Q-learning algorithm to collect training data for ANNs.…”

“…Deep Q-Networks (DQN) is generated by combining the ANNs and the Q-learning algorithms. The preliminary studies of employing the DQN in AGVs route planning on square grids world, showed that the DQN can perform robustly on square grids world [133,136,137].…”

“…Zhang et al [148] employed the Q-learning algorithm to solve the multi-AGV route-planning problem in urban roads networks scenarios, and they presented that the optimal route may not always be available if an unknown emergency occurs. Panov et al [133] proposed a multi-AGV fleet coordination algorithm in smart city environments, they combined Convolutional Neural Networks (CNNs) and DRL to make AGVs fulfill tasks in large environments, and they employed the Transfer Learning (TL) to balance multiple objectives. We noticed the paper because the urban scenarios are similar to warehouses.…”

Delta Schema Network in Model-based Reinforcement Learning

“…From the point of view of creating AGI systems, the DSN algorithm can be used to automatically generate scripts for the behavior of an intelligent agent. These scripts can be used to speed up the agent's own behavior planning process [9,7], or to predict user behavior in a cognitive assistant scenario [12].…”

Path Finding for the Coalition of Co-operative Agents Acting in the Environment with Destructible Obstacles

“…This is similar to integration of task and motion planning [6], but unlike other researchers in this field we do not concentrate on task planning with grasping the objects, which is a typical scenario, but rather on task planning with path finding. The approach we suggest can be of particular interest to solving so-called smart relocation tasks [11], [12] when the mission can not be accomplished without the robots helping each other.…”