Mehmet Mutlu

Where to place cameras on a snake robot: Focus on camera trajectory and motion blur

Visual information is heavily used in robotics, in particular for SLAM applications. Visual SLAM algorithms depend on robust feature extraction and reliable state estimation. Quality of the visual information highly depends on how that information is captured. The nature of snake robots locomotion presents considerable challenges on the quality of images captured by an onboard mobile camera. Although placing the camera on the head of the snake robot has advantages when the robot is stationary since the body can be used as a manipulator observing for the environment, how to place the camera in order to capture more useful images for navigation during locomotion is not clear. In this paper, we present a comparative study to discuss implications of the camera location on field coverage and types of image quality for three snake gaits: Rolling, sidewinding and linear progression. Camera pose during locomotion is examined in detail and quality of images are quantified using a motion blur metric which relates camera egomotion to blur. Linear progression is found to be very promising in terms of supplying sharper images. But, there are also other merits that can be exploited in different locomotion types and camera locations.

Compliant snake robot locomotion on horizontal pipes

In this paper we introduce a body-compliant Modular Snake Robot executing rolling gaits on different cylindrical geometries. In the state of the art it is considered that an active shape adaptation to the terrain while a gait is executed produces better performances than a simple pre-programmed stiff motion without feedback. Several attempts to reproduce such behaviors in snake robots range from compliant shape controllers (acting in joint space) to torque control strategies of elastic actuated joints. In our proposal, we incorporate compliant elements in a modular snake robot structure to passively adapt the robots shape to the environment. The gait control remains simple by acting directly in the robots joint space with known gait generation schemes. To validate our results we performed experiments with compliant modular snake robots rolling on pipes with different geometry characteristics such as different diameters, smooth surfaces, surfaces with presence of obstacles (terrain bumps), and considerable changes in diameter in a single robot run. We evaluated the performance across different robots body-compliance values, measuring the speed of locomotion as well as the power consumption. Our results show that providing a good selection of body compliant elements is a way to maintain high locomotion performance (at least while rolling on pipes) without including additional complex control artifacts to the simple open-loop cyclic gait controller.

A Real-time Inertial Motion Blur Metric: Application to Frame Triggering Based Motion Blur Minimization

Mobile robots suffer from sensory data corruption due to body oscillations and disturbances. In particular, information loss on images captured with onboard cameras can be very high, and such loss may become irreversible or computationally costly to undo. In this paper, we propose a novel method to minimize average motion blur captured by such mobile visual sensors. To this end, we derive a motion blur metric (MMBM) that can be computed in real-time by using only inertial sensor measurements and validate it through comparisons with optic flow computations. The applicability of MMBM is illustrated through a motion blur minimizing system implemented on the SensoRHex hexapod robot by externally triggering an onboard camera based on MMBM values computed in real-time while the robot is walking straight on a flat surface. The resulting motion blur is compared to motion blur levels obtained with a regular, fixed frame-rate image acquisition schedule by both qualitative inspection and using a blind blur metric on captured images. MMBM based motion blur minimization system not only reduces average motion blur, but also avoids frames with extreme motion blur before an image gets corrupted by appropriately delaying the triggering of frame acquisition.

A comparative evaluation of adaptive and non-adaptive Sliding Mode, LQR & PID control for platform stabilization

During the uniform locomotion of compliant legged robots and other terrain vehicles, the body of the robot often exhibits complex oscillations which may have a disturbing effect on onboard sensors. For a camera mounted on such a robot, due to perspective projection, the effects of angular disturbances are particularly pronounced as compared to translational disturbances. This paper is motivated by the particular problem of legged robots exhibiting angular body motions and attempts to evaluate the performance of baseline and state-of-the-art controllers for compensating this undesired motion. For this comparative evaluation, a simplified planar camera platform is considered in a Matlab-Simulink based simulation environment but motion disturbances are collected on a physical experimental robot platform. Although the full stabilization problem is in 3D with three independent axes of rotation, we currently consider a planar case on the pitch axis with a kinematic structure very similar to many parallel actuated 3D platforms. We believe that despite the simplified analysis, the presented performance evaluation provides significant insight into the general problem. The work consist of the derivation of the planar platform model followed by the implementation and comparative testing of 4 different controllers, namely Proportional-Integral-Derivative (PID), Linear Quadratic Regulator (LQR), Sliding Mode (SMC) and Adaptive Sliding Mode (ASMC) controllers. Experimental setup, disturbance collection and finally, the controller performance test results are presented and discussed.

Lagrangian based mathematical modeling and experimental validation of a planar stabilized platform for mobile systems

Typical operating conditions for mobile sensor systems, and in particular mobile robots, exhibit a wide range of mechanical disturbances due their ego-motion. Sensor systems mounted on these mobile platforms often suffer to varying degrees from these disturbances. The quality of acquired data is degraded as a result. For instance, the quality of captured video frames from an onboard camera greatly depends on the angular velocity of the body on which the camera is mounted. Motion blur degradation results if large angular motions are present. In order to compensate for such disturbances, stabilization platforms are used. A common approach is measuring body movements using inertial sensors and attempting their cancellation with actuators and control systems. Design of high performance control systems often requires analytical system models. In this article, a planar stabilization platform is considered, to develop and study its kinematic and simple-to-complex dynamic model. The mathematical derivation of the model is presented with and without neglect of the actuator mass components as well as friction effects. This is followed by the comparative validation of these model alternatives against a realistic numerical model fitted to physical experimental data. The results demonstrate that the analytical model, in particular with the actuator mass and friction components included, provides a high degree of fit to the actual behavior.

Compliant universal grippers as adaptive feet in legged robots

ABSTRACT This work investigates the usage of compliant universal grippers as a novel foot design for legged locomotion. The method of jamming of granular media in the universal grippers is characterized by having two distinct states: a soft, fluid-like state which in locomotion can be used to damp impact forces and enable passive shape adaptation especially on rough terrain, and a hard, solid-like state that is more suited to transmit propulsion forces. We propose a system that actively uses and switches between both states of a foot design based on granular jamming and detail the implementation on a quadruped robotic platform. The mechanism is inspired by the stiffness varying function of the tarsal bones in a human foot, and our aim is to understand how the change of foot stiffness can be used to improve the locomotion performance of legged robots. Using the same open loop trot gait in all experiments, it is shown that a fast state transition enables the robot to profit from both states, leading to more uniform foot placement patterns also on rough terrain compared to other tested feet. This results in overall faster gaits and even enables the robot to climb steeper inclined surfaces. GRAPHICAL ABSTRACT

Effects of passive and active joint compliance in quadrupedal locomotion

ABSTRACT Compliance of the body has a crucial role on locomotion performance. The levels and the distribution of compliance should be well tuned to obtain efficient gait. The leg stiffness changes significantly even during different phases of a single gait cycle. This paper presents an experimental study on different passive and active limb compliance configurations. Each configuration is tested on flat, rough and inclined-rough surfaces, to analyze locomotion performance in diverse conditions. As the active compliance mechanism, Tegotae-based control is selected. Even though active compliance is not its primary use, we show that the Tegotae rule presents intriguing features that have potential to boost gait performance in various scenarios. GRAPHICAL ABSTRACT

Self-reconfigurable modular robot interface using virtual reality: Arrangement of furniture made out of roombots modules, 2017 26th IEEE International Symposium on Robot and Human Interactive Communication (RO-MAN)

Self-reconfigurable modular robots (SRMR) offer high flexibility in task space by adopting different morphologies for different tasks. Using the same simple module, complex and more capable morphologies can be built. However, increasing the number of modules increases the degrees of freedom (DOF) of the system. Thus, controlling the system as a whole becomes harder. Indeed, even a 10 DOFs system is difficult to consider and manipulate. Intuitive and easy to use interfaces are needed, particularly when modular robots need to interact with humans. In this study we present an interface to assemble desired structures and placement of such structures, with a focus on the assembly process. Roombots modules, a particular SRMR design, are used for the demonstration of the proposed interface. Two non-conventional input/output devices — a head mounted display and hand tracking system — are added to the system to enhance the user experience. Finally, a user study was conducted to evaluate the interface. The results show that most users enjoyed their experience. However, they were not necessarily convinced by the gesture control, most likely for technical reasons.

Active stabilization of a stiff quadruped robot using local feedback, 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

Animal locomotion exhibits all the features of complex non linear systems such as multi-stability, critical fluctuation, limit cycle behavior and chaos. Studying these aspects on real robots has been proved difficult and therefore results mostly rely on the use of computer simulation. Simple control approaches — based on phase oscillators — have been proposed and exhibit several of these features. In this work, we compare two types of controllers: (a) an open loop control approach based on phase oscillators and (b) the Tegotae-based closed loop extension of this controller. The first controller has been shown to exhibit synchronization features between the body and the controller when applied to a quadruped robot with compliant leg structures. In this contribution, we apply both controllers to the locomotion of a stiff quadruped structure. We show that the Tegotae-controller exhibits self-organizing behavior, such as spontaneous gait transition and critical fluctuation. Moreover, it exhibits features such as the ability to stabilize both asymmetric and symmetric morphological changes, despite the lack of compliance in the leg.

Natural user interface for lighting control: Case study on desktop lighting using modular robots

Roombots (RB) are self-reconfigurable modular robots designed to explore physical structure change by robotic reconfiguration and adaptive locomotion on structured grid environments or unstructured environments. The primary goal of RB is to create adaptive furniture. In this study, we propose a novel and user-friendly interface to control position and intensity of a mobile desk light using RB modules. In the proposed method, the user interacts with the RB with only hand/arm gestures. The users arm is tracked with a single Kinect having birds eye view. We demonstrate the effectiveness of the proposed interface in real hardware setup and discuss contributions of it.