Semantic simulation engine provides tools to implement mobile robot simulation based on real data delivered by robot observations In the paper the semantic simulation engine and its role in multi level mobile robot operator training tool is described. The PhysX model is used to perform an inspection intervention mobile robot simulation in which collision detection and rigid body simulation is avail- able. This raw data is transformed into a semantic map and then consequently NVIDIA PhysX model is constructed. Semantic map is generated based on raw data acquired by a mobile robot in INDOOR environment. We are presenting a new approach that uses semantic mapping to generate physical model of an environment that is later integrated with mobile robot simulator. There are several robotic applications that use semantic information to build complex environment maps with labeled entities, therefore advanced robot behavior based on ontological information can be processed in high conceptual level. Semantic mapping is still an open problem because of its complexity. The virtual scene composed by several entities and relations between them is the core of the proposed semantic simulation engine. Data is acquired by a mobile robot equipped with 3D measurement system. The paper concerns the research related to the semantic mapping ap- plication for automatic virtual scene generation. In order to study their characteristics, experiments in Subsumption architecture and motor schema are example of their methods. Kata kunci: behavior based robotics, coordination, reinforcement learning, navigasi otonom Behaviors coordination is one of keypoints in behavior based robotics. Sebagai hasilnya, robot telah berhasil melakukan navigasi otonom meski dengan beberapa keterbatasan akibat peletakan dan karakteristik sensor. Algoritma Q learning diterapkan pada subsumption architecture dari suatu robot fisik. Variabel laju pembelajaran berpengaruh pada performa robot dalam fase pembelajaran. Q learning adalah metode pembelajaran reinforcement yang populer digunakan pada pembelajaran robot karena sederhana, konvergen dan off policy. Perilaku yang mampu belajar dapat memperbaiki performa robot dalam menghadapi ketidakpastian. Sedang metode ke dua memberikan respons yang lebih lambat namun lebih halus, dan cenderung menemukan target lebih cepat. Dari hasil eksperimen dapat disimpulkan bahwa metode pertama memberikan respons yang cepat, robust tetapi tidak halus. Untuk mempelajari sifat keduanya, eksperimen pada robot fisik perlu dilakukan. Arsitektur subsumption dan motor schema adalah contoh dari metode tersebut. Koordinasi perilaku adalah salah satu faktor kunci pada robot berbasis perilaku. Finally, we implemented Move-and-Improve using ROS and deployed it on real robots. We also demonstrated through MATLAB simulations the benefits of using our decentralized approach as compared to a centralized Genetic Algorithm approach to solve the MD-MTSP problem. To validate the efficiency of the Move-and-Improve distributed algorithm, we first conducted extensive simulations using Webots and evaluated its performance in terms of total traveled distance, maximum tour length, and ratio of overlapped targets, under different settings. Our approach consists of four main phases: (1) initial target allocation, (2) tour construction, (3) negotiation of conflicting targets, (4) solution improvement. The concept is simple: in each step, a robot moves and attempts to improve its solution while communicating with its neighbors. It involves the cooperation of the robots to incrementally allocate targets and remove possible overlap. This paper provides a distributed solution based on a market-based approach, called Move-and-Improve. As an NP-Hard problem, centralized approaches using meta-heuristic search are typically used to solve it, but such approaches are computation-intensive and cannot effectively deal with the dynamic nature of the system. This problem is known as multi-robot patrolling and can be cast to the multiple depot multiple traveling salesman problem (MD-MTSP), which applies to several mobile robots applications. Consider the problem of having a team of cooperative and autonomous robots to repeatedly visit a set of target locations and return back to their initial locations.
0 kommentar(er)
