International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems Vol. 26, No. 5 (2018) 809–838  World Scientific Publishing Company DOI: 10.1142/S0218488518500368 809 On-Line Fault-Tolerant Fuzzy-Based Path Planning and Obstacles Avoidance Approach for Manipulator Robots Abderraouf Maoudj *,†,‡ , Abdelfetah Hentout † , Brahim Bouzouia * and Redouane Toumi * * University of Sciences and Technology Houari Boumediene (USTHB), Laboratory Robotics Parallelism and Embedded Systems (LRPE), BP 32 El Alia, Bab Ezzouar, Algiers 16111, Algeria † Centre de Développement des Technologies Avancées (CDTA), Division Productique et Robotique (DPR), BP 17, Baba Hassen, Algiers 16303, Algeria ‡ amaoudj@cdta.dz Received 22 January 2017 Revised 3 May 2017 Manipulator robots are widely used in many fields to replace humans in complex and risky environments. However, in some particular environments the robot is prone to failure, resulting in decreased performance. In such environments, it is extremely difficult to repair the robot which interrupts the execution process. Therefore, fault tolerance plays an important role in industrial manipulators applications. In this paper, the key problems related to fault-tolerance and path planning of manipulator robots under joints failures are handled within an on-line fault-tolerant fuzzy-logic based path planning approach for high degree-of-freedom robots. This approach provides an alternative to using mathematical models to control such robots, and improves tolerance to certain faults and mechanical failures. The controller consists of two fuzzy units (i) the first unit, Fuzzy_Path_Planner, is responsible of path planning; (ii) the second unit, Fuzzy_Obstacle_ Avoidance, is conceived for obstacles avoidance. Moreover, the proposed approach is capable of repelling the manipulator away from both local minima and limit cycle problems. Finally, to validate the proposed approach and show its performances and effectiveness, different tests are carried out on two six degree-of-freedom manipulator robots (ULM and PUMA560 robots), accomplishing point- to-point tasks, with and without considering some joints failures. Keywords: Fuzzy logic; manipulator robots; on-line path planning; local minima and limit-cycle problems; obstacles avoidance; mechanical failures. 1. Introduction Today, manipulator robots have occupied a considerable attention for automation of many complex tasks in many fields such as material handling, welding, and electronic assembly in manufacturing system, 1 space maintenance missions, 2 etc. These tasks include a primary problem of getting a manipulator to move safely from an initial position to a Target. The most desired characteristics of these robots are the ability to become autonomous, fully independent and intelligent to deal with any unexpected events during the tasks execution process. To this regard, fault-tolerant path planning is the key to all robotic approaches, as it allows finding a collision-free path to move safely Int. J. Unc. Fuzz. Knowl. Based Syst. 2018.26:809-838. Downloaded from www.worldscientific.com by 198.46.216.91 on 08/08/19. Re-use and distribution is strictly not permitted, except for Open Access articles.