Bronchoscopic diagnosis and intervention in the lung is a new frontier for steerable needles, where they have the potential to enable minimally invasive, accurate access to small nodules that cannot be reliably accessed today. However, the curved, flexible bronchoscope requires a much longer needle than prior work has considered, with complex interactions between the needle and bronchoscope channel, introducing new challenges in steerable needle control. In particular, friction between the working channel and needle causes torsional windup along the bronchoscope, the effects of which cannot be directly measured at the tip of thin needles embedded with 5 degree-of-freedom magnetic tracking coils. To compensate for these effects, we propose a new torsional deadband-aware Extended Kalman Filter to estimate the full needle tip pose including the axial angle, which defines its steering direction. We use the Kalman Filter estimates with an established sliding mode controller to steer along desired trajectories in lung tissue. We demonstrate that this simple torsional deadband model is sufficient to account for the complex interactions between the needle and endoscope channel for control purposes. We measure mean final targeting error of 1.36 mm in phantom tissue and 1.84 mm in ex-vivo porcine lung, with mean trajectory following error of 1.28 mm and 1.10 mm, respectively.
Despite its proven success in a wide variety of applications, the atomic force microscope (AFM) remains limited by its slow imaging rate. One approach to overcome this challenge is to rely on algorithmic approaches that reduce the imaging time not by scanning faster but by scanning less. Such schemes are particularly useful on older instruments as they can provide significant gains despite the existing (slow) hardware. At the same time, algorithms for sub-sampling can yield even greater improvements in imaging rate when combined with advanced scanners that can be retrofitted into the system. In this work, we focus on the use of a dual-stage piezoelectric scanner coupled with a particular scanning algorithm known as Local Circular Scan (LCS). LCS drives the tip of the AFM along a circular trajectory while using feedback to center that circle on a sample edge and to move the circle along the feature, thus reducing imaging time by concentrating the samples to the region of interest. Dual-stage systems are well-suited to LCS as the algorithm is naturally described in terms of a high-frequency, short range path (the scanning circle) and a slower, long range path (the track along the sample). However, control of the scanner is not straightforward as the system is multi-input, single-output. Here we establish controllability and observability of the scanning stage, allowing us to develop individual controllers for the long-range and short-range actuators through the principle of separation. We then use an internal model controller for the short range actuator to track a sinusoidal input (to generate the circular motion) and a state-space set-point tracking controller for the long range actuator. The results are demonstrated through simulation.
This paper presents the design of a series elastic actuator and a higher level controller for said actuator to assist the motion of a user's hand in a linkage based hand exoskeleton. While recent trends in the development of exoskeleton gloves has been to exploit the advantages of soft actuators, their size and power requirements limit their adoption. On the other hand, a series elastic actuator can provide compliant assistance to the wearer while remaining compact and lightweight. Furthermore, the linkage based mechanism integrated with the SEA offers repeatability and accuracy to the hand exoskeleton. By measuring the user's motion intention through compression of the elastic elements in the actuator, a virtual dynamic system can be utilized that assists the users in performing the desired motion while ensuring the motion stability of the overall system. This work describes the detailed design of the actuator followed by performance tests using a simple PD controller on the integrated robotic exoskeleton prototype. The performance of the proposed high level controller is tested using the integrated exoskeleton glove mechanism for a single finger, using two types of input motion. Preliminary results are discussed as well as plans for integrating the proposed actuator and high level controller into a complete hand exoskeleton prototype to perform intelligent grasping.
Having unified representations of human walking gait data is of paramount importance for wearable robot control. In the rehabilitation robotics literature, control approaches that unify the gait cycle of wearable robots are more appealing than the conventional approaches that rely on dividing the gait cycle into several periods, each with their own distinct controllers. In this article we propose employing algebraic curves to represent human walking data for wearable robot controller design. In order to generate algebraic curves from human walking data, we employ the 3L fitting algorithm, a tool developed in the pattern recognition literature for fitting implicit polynomial curves to given datasets. For an impedance model of the knee joint motion driven by the hip angle signal, we provide conditions by which the generated algebraic curves satisfy a robust relative degree condition throughout the entire walking gait cycle. The robust relative degree property makes the algebraic curve representation of walking gaits amenable to various nonlinear output tracking controller design techniques.
Trajectory tracking robotic systems require complex control procedures that occupy less space and need less energy. For these reasons, the development of computerized and integrated control systems is crucial. Recently, developing reconfigurable Field Programmable Gate Arrays (FPGAs) give a prominence of the complete robotic control systems. Furthermore, it has been found in the literature that the model-based control methods are most efficient and cost-effective. This model must interpret how multiple moving parts interact with each other and with their environment. On the other hand, MultiBody Dynamic (MBD) approach is considered to solve these difficulties to attain the models accurately. However, the obtained equations of motion do not match the well-developed forms of control theory. In this paper, the MBD model of a mobile robot is established; and the equations of motion are reshaped into their control canonical form. Additionally, the Sliding Mode Control (SMC) theory is used to design the control law. The constraints' manifold, which is available in the equations of the MBD system, are imposed systematically as the switching surface. SMC is applied because of its ability to address multiple-input/multiple-output nonlinear systems without resorting any approximations. Eventually, the experimental verification of the proposed algorithm is carried out using DaNI mobile robot in which, a Reconfigurable Input/Output (RIO) board is used to reorient the control design, so that can fit the required trajectory. The control law is implemented using LabVIEW software and NI-sbRIO-9631 with acceptable performance. It is obvious that the integration of MBD/SMC/FPGA can be used successfully to develop embedded systems for the applications of trajectory tracking robotics.
Fully automatic parking (FAP) is a key step towards the age of autonomous vehicle. Motivated by the contribution of human vision to human parking, in this paper, we propose a computer vision based FAP method for the autonomous vehicles. Based on the input images from a rear camera on the vehicle, a convolutional neural network (CNN) is trained to automatically output the steering and velocity commands for the vehicle controlling. The CNN is trained by Caffe deep learning framework. A 1/10th autonomous vehicle research platform (1/10-SAVRP), which configured with a vehicle controller unit, an automated driving processor, and a rear camera, is used for demonstrating the parking maneuver. The experimental results suggested that the proposed approach enabled the vehicle to gain the ability of parking independently without human input in different driving settings.
We present a state estimator for a UAV operating in an environment equipped with ultra-wideband radio beacons. The beacons allow the UAV to measure distances to known positions in the world. The estimator additionally uses the vehicle's rate gyroscope and accelerometer, and crucially does not rely on any knowledge of the vehicle's dynamic properties (e.g. mass, mass moment of inertia, aerodynamic properties). This makes the estimator especially useful in situations where the exact system parameters are unknown (e.g. due to unknown payloads), or where the environment is unpredictable (e.g. wind gusts). Experimental results demonstrate the approach's efficacy, and demonstrate that the estimator can run on low-cost microcontrollers with typical sensors.
Underwater robots with buoyancy control capability are highly desirable in deep ocean exploration for underwater environment monitoring and intelligent collection. In this paper, a prototype of buoyancy control device powered by ionic polymer metal composite (IPMC) is developed. An IPMC is used for enhancing the water electrolysis of tap water and separating the gases produced. The produced hydrogen and oxygen gases are stored in two separate chambers. Collection of these gases increase the volume of water displaced by the device, hence, increases its buoyancy. Two solenoid valves are used to control the release of gases to decrease the device’s buoyancy. Using a dynamic model developed in our previous work, the parameters of the model are identified through an open-loop test. A PID controller is then designed for close-loop depth control. The PID controller uses the error in depth to estimate the desired gas generation/releasing rate. It then calculates the duty cycle of the pulse-width modulation (PWM) signal used for driving the solenoid valves. The closed-loop depth control is verified both through simulation and real-time experiment, showing satisfactory results.
This paper introduces a decentralized approach of collaborative control between multiple robots. A dynamic problem is considered to illustrate the effectiveness of this approach. The objective of this problem is to control three robots that are connected to a ball through elastic strings to bring the ball to a pre-defined target position. Since there is no communication between the robots, each robot does not know how the other robots are going to react at any instant. The only information available to the robots are the current and target positions of the ball. Genetic Fuzzy Systems (GFSs) are used to develop controllers for individual robots to tackle this problem. The nonlinearity of fuzzy logic systems coupled with the search capability of Genetic Algorithm (GA) provides an invaluable tool to design controllers for such tasks. The system is first trained through a set of scenarios and then applied to an extensive test set to test the effectiveness of the approach.
Two-phase cooling systems provide a viable technology for high–heat flux rejection in electronic systems. They provide high cooling capacity and uniform surface temperature. However, a major restriction of their application is the critical heat flux condition (CHF). This work presents model predictive control (MPC) design for CHF avoidance in two-phase pump driven cooling systems. The system under study includes multiple microchannel heat exchangers in series. The MPC controller performance is compared to the performance of a baseline PI controller. Simulation results show that while both controllers are able to maintain the two-phase cooling system below CHF, MPC has significant reduction in power consumption compared to the baseline controller.
Collective behavior emerges from local interactions in a group, has been observed in many natural systems, and is of significant interests for engineering applications. The Vicsek model is a mathematical tool to study collective alignment in a group of self-propelled particles based on local interaction, which has been well-studied in the literature for its simple algorithm and complex global behaviors. Several studies show that particles reach alignment faster when the directionality of particle interaction is restricted by an optimal view angle. This result seems counterintuitive, since each particle is expected to get more information through omnidirectional interaction. This work seeks to explore the possible causes of this optimal view angle by studying interaction dynamics in Vicsek model with restricted view angle through numerical simulation.
In this work, we address the problem of deploying a multi robot system for row crop phenotyping. We propose a sampling based navigation algorithm for the team of robots to estimate the underlying spatial distribution in a field. We use Gaussian Process Models to predict the distribution of a scalar function in a field, and choose Mutual Information as a metric for selecting the future samples. With a row crop structure, we present a collision-free assignment and scheduling algorithm for the robots to reach the goal positions which minimizes the total traveling distance. The effectiveness of the proposed algorithm is demonstrated through simulations.
A faulty sensor may lead to degraded system performance, unstable system, or even a fatal accident. On the other hand, the increasing need for safety and reliability has motivated the development of fault-tolerant control (FTC) techniques. This paper proposes a robust fault-tolerant gain-scheduled noisy output-feedback controller (GSNOF) that guarantees system stability and performance in the presence of sensor aging under control input constraints, where the sensor performance degradation due to aging is modeled by its measurement noisy covariance. The closed-loop system stability and performance, in terms of numerical complexity, computation time, and ℋ2 performance, are studied. The proposed controller is compared in simulations with the published results, which shows that the proposed controller is capable of guaranteeing the stability, performance with reduced numerical complexity and computation load under gradual sensor performance degradation, and it is feasible for real-time control.
Low-cost pressure sensors are used for various purposes, but it have unique properties such as hysteresis characteristics and creep characteristics. Therefore, it should be calibrated according to the purpose of use. For this purpose, a calibration device capable of applying a precise and repetitive load was generated. The core performance required for the calibration equipment is force control which can accurately apply the target load. However, since the material that contacts the end-effector of the equipment is a rigid body, it is difficult to control due to its high-speed response characteristics, so a robust and stable sliding mode force controller have to be applied. To design the sliding mode force controller of the calibration equipment, dynamic model of the system was derived, and system identification was performed. In order to improve the performance of the system identification, system model was modified by applying the nonlinear model to the system similar to the Hammerstein model. We applied a nonlinear model compensator to the system model and designed sliding mode force control. To verify the performance of the designed controller, force control was applied to static and dynamic loads. It is confirmed through the verification experiment that the proposed sliding mode force controller has sufficient control performance.
This study presents implementation of an object fetching task with a mobile robotic platform (PR2) for hospital patients. The primary goal of this study is to define the requirements for a robotic nursing assistants. The designed application scenario consists of PR2 robotic platform, human subject as the patient, and a tablet for patient-robot communication. The PR2 robot understands patient's request and fetches the requested object by performing automated action steps. For this designed scenario, human subject experiments are performed. Furthermore, all human subject test results are presented and observations during tests are provided. These activities are part of a larger effort to establish adaptive robotic nursing assistants for physical tasks in hospital environments.
The partitioned adaptive control and vibration suppression of free-floating space robot with flexible arms in post-impact process are studied. At first, the dynamic model of combination system after flexible space robot system capturing a target system is established based on the collision theory; the impact effect of space robot combination system after capture operation is analyzed at the same time. Secondly, based on the double time scale decomposition theory, the unstable combination system is decomposed into fast system and slow system, representing the rigid motion of the system and the flexible vibration respectively. To satisfy the compute capacity of space-borne computer and modular design concept, the slow system is considered as a set of interconnected subsystems and a decentralized adaptive neural network control scheme is designed. Neural network is applied to approximating the unknown dynamic of the subsystems; an adaptive sliding mode controller is designed to eliminate both interconnection term and approximation error. The control algorithm has a cutting edge in independent control signal and reduced calculation amount. The Linear Quadratic Optimal control scheme is designed for fast system to suppress the elastic vibration of the flexible manipulators. At last, numerical example demonstrates the validity of the proposed composite control scheme.
In this paper, we address the decentralized collaborative trajectory planning and target surrounding of multiple Unmanned Air Vehicles (UAVs) in three-dimensional space using Partial Differential Equation (PDE) method. The mission objective is simultaneously arrival of UAVs with safe flight trajectory to a certain radius of an a – priori target. Then by reforming the configuration of swarm, UAVs would circle around the target. The assumption in this work is that the arrival time between the UAVs’ final and initial positions are defined a – priori. The constraints in this paper are (i) Three dimensional Dubins path and UAV dynamic constraints, (ii) Minimum separation distance between UAVs, and (iii) Collision-free trajectory throughout the flight. We define a novel concept of Prediction Set (PS) based on our previous study on PDE path planning method and then we apply the PDE PSs to the constraints of the problem (i.e., (i) to (iii)) and solve the optimization problem. Finally, the concept is demonstrated by numerical simulation and an experiment to represent the effectiveness of the solution.
The filtered basis functions (FBF) approach is gaining interest for feedforward tracking control of linear, especially, nonminimum phase systems. It expresses the control input to the plant as a linear combination of basis functions with unknown coefficients. The basis functions are forward filtered through the plant dynamics and the coefficients are selected such that the tracking error is minimized. This article proposes a robust FBF approach for tracking control of linear time invariant systems with known uncertainty. A robust filter is formulated as the inverse of an optimal controller that minimizes a frequency-domain cost function over the known uncertainty. The proposed robust FBF approach filters the basis functions using the robust filter in lieu of the nominal plant dynamics. Stability issues associated with the robust filter are discussed, as are the incorporation of dynamic uncertainty into the robust filter. Applied to a vibration-prone desktop 3-D printer with dynamic uncertainty, significant improvements in tracking accuracy are demonstrated using the robust FBF approach compared to the standard FBF approach.
The paper presents a study on brain computer interface (BCI) using motor imagery (MI) and facial expressions to control a mobile robot. Traditionally, only MI signals are used in BCI applications. In this paper a hybrid approach of using both MI and facial expression stimulations for BCI is proposed. Electroencephalography (EEG) signals were acquired using a sensor system and processed for several MI and facial expressions to extract characteristic features. The features were used to train support vector machine (SVM) based classifiers and the trained classifiers were used to recognize test signals for correct identification of MI and facial expressions. A system was developed to implement the BCI using MI and facial expressions to control a mobile robot. Results of robot control using MI and facial expressions, individually and together are presented for comparison. The combined features from MI and facial expression stimulations were found to give better performance than MI and facial expressions used individually.
Safety and efficiency are two key elements for planning and control in autonomous driving. Theoretically, model-based optimization methods, such as Model Predictive Control (MPC), can provide such optimal driving policies. Their computational complexity, however, grows exponentially with horizon length and number of surrounding vehicles. This makes them impractical for real-time implementation, particularly when nonlinear models are considered. To enable a fast and approximately optimal driving policy, we propose a safe imitation framework, which contains two hierarchical layers. The first layer, defined as the policy layer, is represented by a neural network that imitates a long-term expert driving policy via imitation learning. The second layer, called the execution layer, is a short-term model-based optimal controller that tracks and further fine-tunes the reference trajectories proposed by the policy layer with guaranteed short-term collision avoidance. Moreover, to reduce the distribution mismatch between the training set and the real world, Dataset Aggregation is utilized so that the performance of the policy layer can be improved from iteration to iteration. Several highway driving scenarios are demonstrated in simulations, and the results show that the proposed framework can achieve similar performance as sophisticated long-term optimization approaches but with significantly improved computational efficiency.
