Jonas Degrave

Terrain Classification for a Quadruped Robot

Using data retrieved from the Puppy II robot at the University of Zurich (UZH), we show that machine learning techniques with non-linearities and fading memory are effective for terrain classification, both supervised and unsupervised, even with a limited selection of input sensors. The results indicate that most information for terrain classification is found in the combination of tactile sensors and proprioceptive joint angle sensors. The classification error is small enough to have a robot adapt the gait to the terrain and hence move more robustly.

Towards Improved Design and Evaluation of Epileptic Seizure Predictors

Objective: Key issues in the epilepsy seizure prediction research are (1) the reproducibility of results (2) the inability to compare multiple approaches directly. To overcome these problems, the seizure prediction challenge was organized on Kaggle.com. It aimed at establishing benchmarks on a dataset with predefined train, validation, and test sets. Our main objective is to analyze the competition format, and to propose improvements, which would facilitate a better comparison of algorithms. The second objective is to present a novel deep learning approach to seizure prediction and compare it to other commonly used methods using patient centered metrics. Methods: We used the competitions datasets to illustrate the effects of data contamination. Having better data partitions, we compared three types of models in terms of different objectives. Results: We found that correct selection of test samples is crucial when evaluating the performance of seizure forecasting models. Moreover, we showed that models, which achieve state-of-the-art performance with respect to commonly used AUC, sensitivity, and specificity metrics, may not yet be suitable for practical usage because of low precision scores. Conclusion: Correlation between validation and test datasets used in the competition limited its scientific value. Significance: Our findings provide guidelines which allow for a more objective evaluation of seizure prediction models.

Morphological Properties of Mass–Spring Networks for Optimal Locomotion Learning

Robots have proven very useful in automating industrial processes. Their rigid components and powerful actuators, however, render them unsafe or unfit to work in normal human environments such as schools or hospitals. Robots made of compliant, softer materials may offer a valid alternative. Yet, the dynamics of these compliant robots are much more complicated compared to normal rigid robots of which all components can be accurately controlled. It is often claimed that, by using the concept of morphological computation, the dynamical complexity can become a strength. On the one hand, the use of flexible materials can lead to higher power efficiency and more fluent and robust motions. On the other hand, using embodiment in a closed-loop controller, part of the control task itself can be outsourced to the body dynamics. This can significantly simplify the additional resources required for locomotion control. To this goal, a first step consists in an exploration of the trade-offs between morphology, efficiency of locomotion, and the ability of a mechanical body to serve as a computational resource. In this work, we use a detailed dynamical model of a Mass–Spring–Damper (MSD) network to study these trade-offs. We first investigate the influence of the network size and compliance on locomotion quality and energy efficiency by optimizing an external open-loop controller using evolutionary algorithms. We find that larger networks can lead to more stable gaits and that the system’s optimal compliance to maximize the traveled distance is directly linked to the desired frequency of locomotion. In the last set of experiments, the suitability of MSD bodies for being used in a closed loop is also investigated. Since maximally efficient actuator signals are clearly related to the natural body dynamics, in a sense, the body is tailored for the task of contributing to its own control. Using the same simulation platform, we therefore study how the network states can be successfully used to create a feedback signal and how its accuracy is linked to the body size.

Transfer learning of gaits on a quadrupedal robot

Learning new gaits for compliant robots is a challenging multi-dimensional optimization task. Furthermore, to ensure optimal performance, the optimization process must be repeated for every variation in the environment, for example for every change in inclination of the terrain. This is unfortunately not possible using current approaches, since the time required for the optimization is simply too high. Hence, a sub-optimal gait is often used. The goal in this manuscript is to reduce the learning time of a particle swarm algorithm, such that the robot’s gaits can be optimized over a wide variety of terrains. To facilitate this, we use transfer learning by sharing knowledge about gaits between the different environments. Our findings indicate that using transfer learning new robust gaits can be discovered faster compared to traditional methods that learn a gait for each environment independently.

Comparing trotting and turning strategies on the quadrupedal oncilla robot

In this paper, we compare three different trotting techniques and five different turning strategies on a small, compliant, biologically inspired quadrupedal robot, the Oncilla. The locomotion techniques were optimized on the actual hardware using a treadmill setup, without relying on models. We found that using half ellipses as foot trajectories resulted in the fastest gaits, as well as the highest robustness against parameter changes. Furthermore, we analyzed the importance of using the scapulae for turning, from which we observed that although not necessary, they are needed for turning with a higher speed.

Developing an embodied gait on a compliant quadrupedal robot

Incorporating the body dynamics of compliant robots into their controller architectures can drastically reduce the complexity of locomotion control. An extreme version of this embodied control principle was demonstrated in highly compliant tensegrity robots, for which stable gait generation was achieved by using only optimized linear feedback from the robots sensors to its actuators. The morphology of quadrupedal robots has previously been used for sensing and for control of a compliant spine, but never for gait generation. In this paper, we successfully apply embodied control to the compliant, quadrupedal Oncilla robot. As initial experiments indicated that mere linear feedback does not suffice, we explore the minimal requirements for robust gait generation in terms of memory and nonlinear complexity. Our results show that a memoryless feedback controller can generate a stable trot by learning the desired nonlinear relation between the input and the output signals. We believe this method can provide a very useful tool for transferring knowledge from open loop to closed loop control on compliant robots.

PLXTRM: Prediction-Led eXtended-guitar Tool for Real-time Music applications and live performance

This article presents PLXTRM, a system tracking picking-hand micro-gestures for real-time music applications and live performance. PLXTRM taps into the existing gesture vocabulary of the guitar player. On the first level, PLXTRM provides a continuous controller that doesn’t require the musician to learn and integrate extrinsic gestures, avoiding additional cognitive load. Beyond the possible musical applications using this continuous control, the second aim is to harness PLXTRM’s predictive power. Using a reservoir network, string onsets are predicted within a certain time frame, based on the spatial trajectory of the guitar pick. In this time frame, manipulations to the audio signal can be introduced, prior to the string actually sounding, ’prefacing’ note onsets. Thirdly, PLXTRM facilitates the distinction of playing features such as up-strokes vs. down-strokes, string selections and the continuous velocity of gestures, and thereby explores new expressive possibilities.

Oncilla Robot: A Versatile Open-Source Quadruped Research Robot With Compliant Pantograph Legs

We present Oncilla robot, a novel mobile, quadruped legged locomotion machine. This large-cat sized, 5.1 kg robot is one of a kind of a recent, bioinspired legged robot class designed with the capability of model-free locomotion control. Animal legged locomotion in rough terrain is clearly shaped by sensor feedback systems. Results with Oncilla robot show that agile and versatile locomotion is possible without sensory signals to some extend, and tracking becomes robust when feedback control is added (Ajallooeian, 2015). By incorporating mechanical and control blueprints inspired from animals, and by observing the resulting robot locomotion characteristics, we aim to understand the contribution of individual components. Legged robots have a wide mechanical and control design parameter space, and a unique potential as research tools to investigate principles of biomechanics and legged locomotion control. But the hardware and controller design can be a steep initial hurdle for academic research. To facilitate the easy start and development of legged robots, Oncilla-robots blueprints are available through open-source. The robots locomotion capabilities are shown in several scenarios. Specifically, its spring-loaded pantographic leg design compensates for overdetermined body and leg postures, i.e., during turning maneuvers, locomotion outdoors, or while going up and down slopes. The robots active degree of freedom allow tight and swift direction changes, and turns on the spot. Presented hardware experiments are conducted in an open-loop manner, with little control and computational effort. For more versatile locomotion control, Oncilla-robot can sense leg joint rotations, and leg-trunk forces. Additional sensors can be included for feedback control with an open communication protocol interface. The robots customized actuators are designed for robust actuation, and efficient locomotion. It trots with a cost of transport of 3.2 J/(Nm), at a speed of 0.63 m s-1 (Froude number 0.25). The robot trots inclined slopes up to 10°, at 0.25 m s-1. The multi-body Webots model of Oncilla robot, and Oncilla robots extensive software architecture enables users to design and test scenarios in simulation. Controllers can directly be transferred to the real robot. Oncilla robots blueprints are open-source published (hardware GLP v3, software LGPL v3).

Dual Rectified Linear Units (DReLUs): A replacement for tanh activation functions in Quasi-Recurrent Neural Networks

Rectified Linear Units (ReLUs) are widely used in feed-forward neural networks, and in convolutional neural networks in particular. However, they can be rarely found in recurrent neural networks due to the unboundedness and the positive image of the rectified linear activation function. In this paper, we introduce Dual Rectified Linear Units (DReLUs), a novel type of rectified unit that comes with a positive and negative image that is unbounded. We show that we can successfully replace the tanh activation function in the recurrent step of quasi recurrent neural networks. In addition, DReLUs are less prone to the vanishing gradient problem, they are noise robust, and they induce sparse activations. Therefore, we are able to stack up to eight quasi recurrent layers, making it possible to improve the current state-of-the-art in character-level language modeling over architectures based on shallow Long Short-Term Memory (LSTM).