A Hybrid Technique for Active SLAM Based on RPPO Model with Transfer Learning

The problem of exploration in unknown environments is still a great challenge for autonomous mobile robots due to the lack of a priori knowledge. Active Simultaneous Localization and Mapping (SLAM) is an effective method to realize obstacle avoidance and autonomous navigation. Traditional Active SLAM is usually complex to model and difficult to adapt automatically to new operating areas. This paper presents a novel Active SLAM algorithm based on Deep Reinforcement Learning (DRL). The Relational Proximal Policy Optimization (RPPO) model with deep separable convolution and data batch processing is used to predict the action strategy and generate the action plan through the acquired environment RGB images, so as to realize the autonomous collision free exploration of the environment. Meanwhile, Gmapping is applied to locate and map the environment. Then, based on Transfer Learning, Active SLAM algorithm is applied to complex unknown environments with various dynamic and static obstacles. Finally, we present several experiments to demonstrate the advantages and feasibility of the proposed Active SLAM algorithm.

Optimized Graph Search Algorithms for Exploration with Mobile Robot

Graph search algorithms and shortest path algorithms, designed to allow real mobile robots to search unknown environments, are typically run in a hybrid manner, which results in the fast exploration of an entire environment using the shortest path. In this study, a mobile robot explored an unknown environment using separate depth-first search (DFS) and breadth-first search (BFS) algorithms. Afterward, developed DFS + Dijkstra and BFS + Dijkstra algorithms were run for the same environment. It was observed that the newly developed hybrid algorithm performed the identification using less distance. In experimental studies with real robots, progression with DFS for the first-time discovery of an unknown environment is very efficient for detecting boundaries. After finding the last point with DFS, the shortest route was found with Dijkstra for the robot to reach the previous node. In defining a robot that works in a real environment using DFS algorithm for movement in unknown environments and Dijkstra algorithm in returning, time and path are shortened. The same situation was tested with BFS and the results were examined. However, DFS + Dijkstra was found to be the best algorithm in field scanning with real robots. With the hybrid algorithm developed, it is possible to scan the area with real autonomous robots in a shorter time. In this study, field scanning was optimized using hybrid algorithms known.