Intelligent Force Control Research Articles

Non-Intelligent and Intelligent Force Control for Robotic Applications: A Review

This paper presents an overview of force control approaches for robotic systems. It covers three main methods: non-intelligent methods, intelligent methods, and recent methods. In each section, the discussion focused on how the researcher implements their methods in control system to obtain the desired force control for system’s robustness towards external disturbances and internal uncertainties. The purpose of applying force control is to ensure that the executed robotic task does not damage the manipulated object or environment. The benefits for each method were highlighted at the end of each section.

Intelligent Adaptive Active Force Control of a Robotic Arm with Embedded Iterative Learning Algorithms

Kawalan jitu dan lasak bagi satu sistem lengan robot atau pengolah adalah amat penting terutama sekali jika sistem mengalami pelbagai bentuk bebanan dan keadaan pengendalian. Kertas kerja ini memaparkan satu kaedah baru dan lasak untuk mengawal lengan robot menggunakan teknik pembelajaran secara berlelaran yang dimuatkan dalam strategi kawalan daya aktif. Sebanyak dua algoritma pembelajaran utama digunakan dalam kajian – yang pertama digunakan untuk menala gandaan pengawal secara automatik manakala yang satu lagi pula untuk menganggarkan matriks inersia pengolah. Kedua-dua parameter ini dihasilkan secara adaptif dan dalam talian ketika robot sedang menjalankan tugas menjejak trajektori dalam persekitaran tindakan daya gangguan. Dalam kajian ini, pengetahuan awal tentang kedua–dua nilai gandaan pengawal dan anggaran matriks inersia tidak wujud. Dengan demikian, suatu skema kawalan yang jitu dan lasak terhasil. Keberkesanan kaedah yang dicadangkan dapat ditentusahkan melalui hasil kajian yang diperoleh dan dibentangkan dalam kertas kerja ini. Kata kunci: Adaptif; kawalan daya aktif; pembelajaran berlelaran; matriks inersia; gandaan pengawal The robust and accurate control of a robotic arm or manipulator are of prime importance especially if the system is subjected to varying forms of loading and operating conditions. The paper highlights a novel and robust method to control a robotic arm using an iterative learning technique embedded in an active force control strategy. Two main iterative learning algorithms are utilized in the study – the first is used to automatically tune the controller gains while the second to estimate the inertia matrix of the manipulator. These parameters are adaptively computed on-line while the robot is executing a trajectory tracking task and subject to some forms of external disturbances. No priori knowledge of both the controller gains and the estimated inertia matrix are ever assumed in the study. In this way, an adaptive and robust control scheme is derived. The effectiveness of the method is verified and can be seen from the results of the work presented in this paper. Keywords: Adaptive; active force control; iterative learning; inertia matrix; controller gain

Intelligent Force Control of a Flip Chip Mounting System

In this paper, we have developed a new mounting head system for flip chip. The proposed head system consists of a macro/micro positioning actuator for stable force control. The macro actuator provides the system with a gross motion while the micro device yields fine tuned motion to reduce the harmful impact force that occurs between very small sized electronic parts and the surface of a PCB(printed circuit board). In order to show the effectiveness of the proposed macro/micro chip mounting system, we compared the proposed system with the conventional chip mounting head equipped with a macro actuator only. A series of experiments were executed under the mounting conditions such as various access velocities and PCB stiffness. As a result of this study, a satisfactory voice coil actuator as the micro actuator has been developed , and its performance meet well the specifications desired for the design of the chip mounting head system and show good correspondence between theoretical analysis and experimental results.

동역학 기반의 지능 힘제어 방식을 이용한 이동 로봇의 장애물 회피에 대한 연구

In this paper, a new collision avoidance algorithm based on the dynamic model of a mobile robot is proposed. In order to avoid obstacles on the path of a mobile robot, intelligent force control is used to regulate accurate distance between a robot and an obstacle. Since uncertainties from robot and environment dynamics degrade the performance of a collision avoidance task, neural network is used to compensate for uncertainties so that the collision avoidance can be performed intelligently. Simulation studies are conducted to confirm the proposed collision avoidance tracking control algorithm.

Enhancing the Autonomy of Teleoperated Redundant Manipulators Through Fusion of Intelligent Control Modules

This paper presents a method for redundancy resolution of an industrial manipulator in a teleoperated force control task. A seven degree-of-freedom (DOF) industrial manipulator manufactured by the Mitsubishi Heavy Industries Ltd. is used for experiments. The task involves obeying a force command sent from a remote computer while autonomously adapting the posture to avoid unexpected obstacles moving toward the manipulator. Redundancy resolution is employed for autonomous adaptation of the configuration to avoid the obstacle while continuing the force control task. This self-adaptive skill on the slave manipulator side is very important because teleoperation is often performed in dangerous or partially unknown environments where unexpected changes such as moving obstacles can well be expected. In such situations, the control ability of the master side is very limited due to the practical limitations of vision sensors to capture a comprehensive view of the environment and the limitations of the degrees of freedom on the master manipulator. The proposed method relies on two modules of an intelligent controller on the slave side. The first is an on-line fuzzy neural network (FNN) for intelligent force control, and the second is a configuration controller that works in harmony with the first to exploit redundancy to react to avoid moving obstacles such that the latter does not inhibit the progress of the former. The second controller generates joint velocity commands in null space of the hand Jacobian, so that its activation does not affect the force controller. Here we show that the proposed method can skillfully avoid a moving obstacle without stopping the force control task. This skillful adaptation ability can significantly improve the efficiency and safety of teleoperated force control tasks with less burden on the master side. This paper presents some promising experimental results to demonstrate the effectiveness of the proposed method.

Neural-fuzzy force control using a TAI-lorable parallel computer concept

In this paper a new control system for the intelligent force control of multifingered robot grips is presented. The multi-level system architecture combines both fuzzy-based and a neural-based control algorithms. After defining the fine manipulation problem fuzzy and neural controller are developed. The demands of flexibility and real-time control of the implementation of the control system are suited by a tailorable parallel computer concept. The basic ideas of the concept are to set up each processing element and the communication individually for its application

Neural Network Control of Grasp Forces for a Robot Hand

In this paper, we present a neural-based force adaptation level of a multi-level system architecture for the intelligent force control of robotic grip systems. The control algorithms developed are implemented by three-layered back-propagation neural networks. A computer based simulation system for the peg-in-hole insertion task is developed to analyze the capabilities of the neural controllers. Their behaviour is discussed by comparing them to conventional and fuzzy-based force controllers performing the same task.

Neural-based learning in grasp force control of a robot hand

In this paper, we present a new control system for the intelligent force control of multifingered robot grips which combines both fuzzy-based adaptation level and a neural-based one with a conventional PID-controller. The most attention is given to the neural-based force adaptation level implemented by three-layered back-propagation neural networks. A computer based simulation system for the peg-in-hole insertion task is developed to analyze the capabilities of the neural controllers. Their behaviour is discussed by comparing them to conventional and fuzzy-based force controllers performing the same task.