RSS 2016 Papers

In this paper we present an inverse optimal control based transfer of motions from human experiments to humanoid robots and apply it to walking in constrained environments. To this end we introduce a 3D template model, which describes motion on the basis of center of mass trajectory, foot trajectories, upper body orientation and phase duration. Despite of its abstract architecture with prismatic joints combined with damped series elastic actuators instead of knees, the model (including dynamics and constraints) is suitable to describe both, human and hu- manoid locomotion with appropriate parameters. We present and apply an inverse optimal control approach to identify optimality criteria based on human motion capture experiments. The iden- tified optimal strategy is then transferred to the humanoid robot for gait generation by solving an optimal control problem, which takes into account the properties of the robot and differences in the environment. The results of this approach are the center of mass trajectory, the foot trajectories, the torso orientation, and the single and double support phase durations for a sequence of multiple steps allowing the humanoid robot to walk within a new environment. We present one computational cycle (from motion capture data to an optimized robot motion) for the example of walking over irregular step stones with the aim to transfer the motion to two very different humanoid robots (iCub Heidelberg01 and HRP-2 14). The transfer of these optimized robot motions to the real robots by means of inverse kinematics is work in progress and not part of this paper.

The advantage of modular robot systems lies in their flexibility, but this advantage can only be realized if there exists some reliable, effective way of generating configurations (shapes) and behaviors (controlling programs) appropriate for a given task. In this paper, we present an end-to-end system for addressing tasks with modular robots, and demonstrate that it is capable of accomplishing challenging multi-part tasks in hardware experiments. The system consists of four tightly integrated components: (1) A high-level mission planner, (2) A large design library spanning a wide set of functionality, (3) A design and simulation tool for populating the library with new configurations and behaviors, and (4) modular robot hardware. The broader goal of this project is enabling users to address real-world tasks using modular robots. We believe this work represents an important step toward this larger goal.

In order to track dynamic objects in a robot’s environment, one must first segment the scene into a collection of separate objects. Most real-time robotic vision systems today rely on simple spatial relations to segment the scene into separate objects. However, such methods fail under a variety of real- world situations such as occlusions or crowds of closely-packed objects. We propose a probabilistic 3D segmentation method that combines spatial, temporal, and semantic information to make better-informed decisions about how to segment a scene. We begin with a coarse initial segmentation. We then compute the probability that a given segment should be split into multiple segments or that multiple segments should be merged into a single segment, using spatial, semantic, and temporal cues. Our probabilistic segmentation framework enables us to significantly reduce both undersegmentations and oversegmentations on the KITTI dataset [3, 4, 5] while still running in real-time. By combining spatial, temporal, and semantic information, we are able to create a more robust 3D segmentation system that leads to better overall perception in crowded dynamic environments.

This paper examines the relationship between sys- tem dynamics and problem complexity of collision avoidance in multi-agent systems. Motivated particularly by results in the field of automated driving, a variant of the reciprocal n-body collision avoidance problem is considered. In this problem, agents must avoid collision while moving according to individual reward functions in a crowded environment. The main contribution of this work is the novel result that there is a quantifiable relationship between system dynamics and the requirement for agent coordination, and that this requirement can change the complexity class of the problem dramatically: from P to NEXP or even NEXPNP. In addition, a constructive proof is provided that demonstrates the relationship and potential real- world applications of the result are discussed.

In robotic systems with moving cameras control of gaze allows for image stabilization, tracking and attention switching. Proper integration of these capabilities lets the robot exploit the kinematic redundancy of the oculomotor system to improve tracking performance and extend the field of view, while at the same time stabilize vision to reduce image blur induced by the robot’s own movements. Gaze may be driven not only by vision but also by other sensors (e.g. inertial sensors or motor encoders) that carry information about the robot’s own movement. Humanoid robots have sophisticated oculomotor systems, usually mounting inertial devices and are therefore an ideal platform to study this problem. We present a complete architecture for gaze control of a humanoid robot. Our system is able to control the neck and the eyes in order to track a 3D cartesian fixation point in space. The redundancy of the kinematic problem is exploited to implement additional behaviors, namely passive gaze stabilization, saccadic movements, and vestibulo- ocular reflex. We implement this framework on the iCub’s head, which is equipped with a 3-DoFs neck and a 3-DoFs eyes system and includes an inertial unit that provides feedback on the acceleration and angular speed of the head. The framework presented in this work can be applied to any robot equipped with an anthropomorphic head. In addition we provide an open- source, modular implementation, which has been already ported to other robotic platforms.

Physiological measures, such as pain, anxiety, effort, or energy consumption, play a crucial role in the evaluation and development of assistive robotic devices. Physiological data are collected and analyzed by researchers and clinicians, and are often used to inform an iterative tuning process of a device and its controller. Currently, these data are collected then analyzed offline such that they are only evaluated after the experiment has ended. This makes any iterative design process tedious and time consuming since tuning must be done on a subject-by-subject basis and for a variety of tasks that the device is intended to be used for. To overcome these drawbacks, we are proposing a new type of human-machine interaction that is based on measuring and using physiological measurements in real-time. By continuously monitoring a physiological objective through a set of suitable sensors, we propose conducting an optimization of a set of controller parameters that shape the assistance provided by the device. In other words, we pose an optimization that includes the human body in the loop. This Body-in-the-Loop optimization allows for optimal subject specific control and has the potential to be used for controller adaptation to changing environments. We validated this concept in an extensive human subject study where we autonomously optimized the actuation onset of a pair of bilateral ankle exoskeletons to minimize user’s metabolic effort.