Intelligent rotational direction control of passive robotic joint with embedded sensors
Highlights
► Passive compliant robotic joint with internal embedded sensors. ► Safe robotic mechanisms with an internally measuring system. ► Control algorithm for detecting the direction of the robotic joint angular rotation. ► Application of conductive silicone rubber in compliant robotic joints. ► Embedded sensors operate as absorbers of excessive collision force.
Introduction
Many research efforts in robotics aim at enabling the robot to safely and robustly navigate and to move about both known and unknown environments. If, however, obstacle avoidance fails, robots often are unable to detect collisions since many designs lack touch sensors. The robot is unaware of the failure of its intended action and ends up in a situation it is unable to resolve and collision may occur. Article (Hoffmann & Gohring, 2005) investigated the possibilities of detecting collisions of a 4-legged robot using the walking engine and software framework. In (Fischer & Henrich, 2009) was presented the first approach that uses multiple 3D depth images for fast collision detection of multiple unknown objects. An exploration method was presented in Ciocarlie (2004) for large robots sparsely equipped with sonar sensors operating in a man shaped outdoors environment, dominated by planar surfaces such as walls or fences. An integrated control framework for safe physical human-robot interaction was presented (De Luca & Flacco, 2012) based on a hierarchy of consistent behaviors. Elastic bands are proposed in Quinlan (1993) as the basis for a new framework to close the gap between global path planning and real-time sensor-based robot control. Subjected to artificial forces, the elastic band deforms in real time to a short and smooth path that maintains clearance from the obstacles.
The joints used in robotics usually include rigid joints and passive compliant joints (Park et al., 2009, Yoon et al., 2003, Yoon et al., 2005). The robots with rigid joints can work with high accuracy, but the contact with other things is normally hard. Passive compliant joints ensure a smooth contact with the surroundings, especially if robots are in contact with humans, but this passive compliant joint cannot determine precisely the position of the members of the joint (Mayer et al., 2006, Shim et al., 2009, Yamano et al., 2003, Yamano et al., 2005). In the robotic joints with compliance, the joints are usually composed of mechanical components such as springs and dampers as internal elements, which absorb the excessive collision force. In many articles, a compliant robotic joint refers to a safe link robotic mechanism which is used in service-robots to prevent hard collision with external objects (Park et al., 2007, Park et al., 2008, Park et al., 2009).
The problem of the use of passive compliant joint is the following: the position of the members of the joint is unknown within the elastic range. To solve this problem a sensor is necessary, which can be made of conductive silicone rubber. Conductive silicone rubber is a material composed of non-conductive elastomer, in which conductive particles are allocated homogeneously (Princy, Joseph, & Kartha, 1998). Since they affect strain in the material by external force, the allocating conditions of the particles are changed to vary the electrical resistance of the material. Due to such property of the material, the compliance, damping and sensory characteristics can be combined in one element. The main task of this study is to investigate the application of conductive silicone rubber in compliant robotic joints not only as a sensing element but also as an element with elastic and damping characteristics. In (Paulick, Bärsch, Lambrecht, & Grunwald, 2008) was introduced that the specific resistivity of the highly conductive filled silicone rubber, which was used there for their investigation shows extensive stability under these load factors electricity, strain and temperature. This compliant robotic joint with sensor-elements of conductive silicone rubber has the benefits of both of the above-mentioned joints, rigid and passive compliant. These sensor-elements are embedded in the joint to detect the direction of the robotic joint angular rotation when external collision force is applied. Besides that, the sensors can operate as absorbers of excessive collision force.
To design control algorithm for detection of rotational direction of the passive robotic joint, fuzzy logic (FL) (Mamdani, 1974, Zadeh, 1965a, Zadeh, 1965b) and artificial neural network (ANN) methods was applied in this article. ANNs are a family of intelligent algorithms which can be used for time series prediction, classification, and control and identification purposes. Neural networks can learn from data. However, understanding the knowledge learned by neural networks has been difficult. Unlike neural networks, however, fuzzy logic by itself cannot learn. It is natural to merge these two techniques and this merged technique is adaptive neuro fuzzy inference system (ANFIS) (Jang, 1993). ANFIS, as a hybrid intelligent system that enhances the ability to automatically learn and adapt, was used by researchers for modeling (Ghandoor and Samhouri, 2009, Nguyen and Ngo, 2008, Singh et al., 2012, Yetilmezsoy et al., 2011), predictions (Hosoz et al., 2011, Khajeh et al., 2009, Paulick et al., 2008, Sivakumar and Balu, 2010) and control (Altin et al., 2012, Areed et al., 2010, Kurnaz et al., 2010, Ravi et al., 2011, Tian and Collins, 2005) in various engineering systems.
In this paper, the application of ANFIS is proposed to detect collisions of the output robotic segments with the external object and further to determine rotational direction of the joint. A control algorithm based on embedded sensors voltage changing was derived to perform the tasks. To evaluate the control algorithm, many experiments were conducted with the robotic joint. Obtained experimental data were used as the ANFIS training and checking data.
Section snippets
Design of robotic joint
A robotic joint design for experimental testing was developed. The main purpose of this model is experimental testing of embedded silicone sensors. The main principle of the joint is shown in Fig. 1. The joint is composed of the two main parts, exterior and interior part. Embedded silicone sensors are positioned between these two parts as shown in the cross-section of this joint in Fig. 1. These sensors have alternate deforming during the angular rotation of the interior part of the robotic
Design of sensors
Due to results of several trial tests and constructive demands, the sensors were developed in the parallelepiped shape (initial length 15 ± 0.1 mm, initial width 9 ± 0.1 mm, and initial height 9 ± 0.1 mm). The sensors were made by press-curing from carbon-black filled silicone rubber (Elastosil® R570/50, Shore A 50) with a special tool. The changing of electrical resistance of the elements was measured with the help of electrodes connected at the bottom and top side of each element. The electrodes were
Working principle of the robotic joint
The designed robotic joint could be used for one of the following tasks, either to detect the angular rotation direction of this joint when an external force prevents its angular rotation, or to detect the collision force direction during angular rotation. Choosing one of these tasks is related to the used application of the joint. The working principle for these tasks is the same and it depends on using the sensors resistance values and one of these two directions to detect another direction.
Experimental setup
LabVIEW virtual instrument and National Instruments BNC-2120 card were used for monitoring and visualizing the results during the movement of the joint. The resistance changes of the sensors were monitored simultaneously during the joint angular rotation and in one diagram visualized. Throughout the test, the time-resistance numerical results were recorded and at the same time visualized. Since measuring resistance in LabVIEW needed a voltage divider, the voltage divider was composed of four
Experimental results
The detection of the robotic arm collision with the external objects was tested by rotating the interior part of the robotic joint as the exterior part of the joint was fixed (in order to illustrate the collision force) and correspondent electrical resistance changing was monitoring which will be the result of simultaneously relaxing two of the embedded sensors and pressing the other two sensors. Fig. 9b shows a situation where sensors 1 and 4 were compressed and sensors 2 and 3 were relaxed.
ANFIS controller design
A controller is a device which controls each and every operation in a decision-making system. From the control system point of view, it brings stability to the system when there is a disturbance, thus safeguarding the equipment from further damage. It may be a hardware-based controller or a software-based controller or a combination of both. In this section, the development of the control strategy for control of the joint angular rotation was presented using the concepts of ANFIS control
Simulation results
The Simulink model for the control of the robotic joint rotational direction was developed in Matlab (Fig. 16). This Simulink model with the ANFIS controller was developed using the various toolboxes available in the Simulink library. The entire system modeled represents an opened loop control system consisting of controllers, samplers, comparators, constants, the mux, gain blocks, constant blocks, ANFIS editor blocks, output sinks (scopes), input sources, etc. The prescribed ANFIS output
Conclusion
The research described in this paper demonstrates that new design of passive compliant robotic joint with embedded sensors can be used for detecting the direction of the robotic joint angular rotation when this joint is exposed to external collision force. That was done by using the sensors made from conductive silicon rubber. So far, there were only embedded springs and dumpers for robot collision safety. The springs and dumpers cannot detect the collision force. In this research the
References (33)
- et al.
DSPACE based adaptive neuro-fuzzy controller of grid interactive inverter
Energy Conversion and Management
(2012) - et al.
Adaptive neuro-fuzzy control of an induction motor
Ain Shams Engineering Journal
(2010) - et al.
An adaptive neuro-fuzzy inference system model for predicting the performance of a refrigeration system with a cooling tower
Expert Systems with Applications
(2011) - et al.
Application of adaptive neuro-fuzzy inference system for solubility prediction of carbon dioxide in polymers
Expert Systems with Applications
(2009) - et al.
Adaptive neuro-fuzzy inference system based autonomous flight control of unmanned air vehicles
Expert Systems with Applications
(2010) - et al.
Estimation of elastic constant of rocks using an ANFIS approach
Applied Soft Computing
(2012) - et al.
Adaptive neuro-fuzzy control of a flexible manipulator
Mechatronics
(2005) - et al.
An adaptive neuro-fuzzy approach for modeling of water-in-oil emulsion formation
Colloids and Surfaces A: Physicochemical and Engineering Aspects
(2011) Fuzzy sets
Information and Control
(1965)- Ciocarlie, M. (2004). Obstacle avoidance and path planning using a sparse array of sonars. New York, NY 10027, USA:...
Electricity consumption in the industrial sector of Jordan: Application of multivariate linear regression and adaptive neuro-fuzzy techniques
Jordan Journal of Mechanical and Industrial Engineering
ANFIS: Adaptive-Network-based Fuzzy Inference Systems
IEEE Transactions on Systems, Man, and Cybernetics
Application of fuzzy algorithms for control of simple dynamic plant
Proceedings of the IEE
Cited by (30)
Piezoresistive sensor fiber composites based on silicone elastomers for the monitoring of the position of a robot arm
2021, Sensors and Actuators, A: PhysicalCitation Excerpt :While encoder systems are sufficient for applications involving stiff robots [4,5], when it comes to soft robotics, rigid sensors can't be used. Soft robotic sensors, like strain sensors, are of interest because they allow monitoring the motion of soft robotic systems for better controlling, which helps to improve the functionality of complex robotic systems [6–8]. The piezoresistive signal can be used to create a closed-loop control system for the actuator.
A generalized dynamic fuzzy neural network based on singular spectrum analysis optimized by brain storm optimization for short-term wind speed forecasting
2017, Applied Soft Computing JournalCitation Excerpt :The adaptive neuro-fuzzy inference system (ANFIS) is a hybrid intelligent system that reinforces the ability of automatic learning and adaptation [62,63]. It is applied by many researchers for modeling [64], control [65] and forecasting [66]. The structure of the approach is composed of five layers:
Robust control by adaptive Non-singular Terminal Sliding Mode
2017, Engineering Applications of Artificial IntelligenceCitation Excerpt :Kalman filter for linear systems and its variation for nonlinear systems such as Extended Kalman Filter (EKF) and Unscented Kalman Filter (UKF), have been one of the most popular state estimation techniques for state estimation purposes (Auger et al., 2013; Simon, 2006; Shabbouei Hagh et al., 2016). Different methods have been developed to sense practical systems’ movements/positions (Petković et al., 2013c; Petkovic et al., 2013b). On the other hand, enhancing the trajectory tracking performance is the other aspect for which considerable efforts have been made to develop new advanced control strategies.
Analyzing of flexible gripper by computational intelligence approach
2016, MechatronicsCitation Excerpt :The signals from embedded sensors were used for estimation of the grasping object shapes and therefore for optimal input displacement prediction. Aiming at ensuring optimal functioning of the gripper, many soft computing techniques are used today such as the fuzzy logic (FL) [47,48], ANN [49], neuro-fuzzy [50,51] and support vector machines (SVM) [52,53]. The basic idea behind the soft computing methodology is to collect input/output data pairs and to learn the proposed network from these data.