ICRA 2015 Papers

What is it that makes movement around obstacles hard? The answer seems clear: obstacles contort the geometry of the workspace and make it difficult to leverage what we consider easy and intuitive straight-line Cartesian geometry. But is Cartesian motion actually easy? It's certainly well-understood and has numerous applications. But beneath the details of linear algebra and pseudoinverses, lies a non-trivial Riemannian metric driving the solution. Cartesian motion is easy only because the pseudoinverse, our powerhouse tool, correctly represents how Euclidean workspace geometry pulls back into the configuration space. In light of that observation, it reasons that motion through a field of obstacles could be just as easy as long as we correctly account for how those obstacles warp the geometry of the space. This paper explores extending our geometric model of the robot beyond the notion of a Cartesian workspace space to fully model and leverage how geometry changes in the presence of obstacles. Intuitively, impenetrable obstacles form topological holes and geodesics curve around them accordingly. We formalize this intuition and develop a general motion optimization framework called Riemannian Motion Optimization (RieMO) to efficiently find motions using our geometric models. Our experiments demonstrate that, for many problems, obstacle avoidance can be much more natural when placed within the right geometric context.

Much research has been done recently on getting various UAVs to perch on various surfaces, however very little research has looked at how to detect when this perch has failed, especially when the surface the UAV is perched on is moving. This paper proposes a method to detect these types of falls using the Instantaneous Center of Rotation (ICR) of the UAV. Two methods are proposed to calculate this ICR, one based on integrating accelerometers to get velocities at various points on the UAV, the other based on using the magnitude of the acceleration at these points to estimate the distance to the ICR from that point. These methods provide a way to detect a fall from a moving perch that should work with different types of perching mechanisms and perches, while requiring minimal additional hardware on the UAV.

Before designing a controller in detail, we ask this question: for a given robot state, and every possible control action, what is the minimum number of steps needed to get to a given target state (if it is possible to do so)? Our biped model is a 2D inverted pendulum with massless legs. We have two controls: (i) the magnitude of an impulsive push-off just before heel-strike and (ii) the step length (location of the next heel-strike). The maximum impulse and minimum step time are bounded to reflect limited motor strength. We compute the set of initial mid-stance velocities from which the biped can reach a given target mid-stance velocity in n or fewer steps. The result: for most target speeds and initial velocities, and with realistically strong actuators, it is possible to reach the target in two steps, if it is possible to reach it at all. This `two steps is enough' result expands on Koolen et al.'s [1] results for capturability of the linear inverted pendulum and is consistent with some human balance and visual guidance experiments.

This paper presents a framework for synchronising multiple triggered sensors with respect to a local clock using standard computing hardware. Providing sensor measurements with accurate and meaningful timestamps is important for many sensor fusion, state estimation and control applications. Accurately synchronising sensor timestamps can be performed with specialised hardware, however, performing sensor synchronisation using standard computing hardware and non-real-time operating systems is difficult due to inaccurate and temperature sensitive clocks, variable communication delays and operating system scheduling delays. Results show the ability of our framework to estimate time offsets to sub-millisecond accuracy. We also demonstrate how synchronising timestamps with our framework results in a tenfold reduction in image stabilisation error for a vehicle driving on rough terrain. The source code will be released as an open source tool for time synchronisation in ROS.

In this paper, we present a novel learning framework for traversable region detection. Firstly, we construct features from the super-pixel level which can reduce the computational cost compared to pixel level. Multi-scale super-pixels are extracted to give consideration to both outline and detail information. Then we classify the multiple-scale super-pixels and merge the labels in pixel level. Meanwhile, we use weighted ELM as our classifier which can deal with the imbalanced class distribution since we only assume that a small region in front of robot is traversable at the beginning of learning. Finally, we employ the online learning process so that our framework can be adaptive to varied scenes. Experimental results on three different style of image sequences, i.e. shadow road, rain sequence and variational sequence, demonstrate the adaptability, stability and parameter insensitivity of our method to the varied scenes and complex illumination.

The ability to safely navigate is a crucial prerequisite for truly autonomous systems. A robot has to distinguish obstacles from traversable ground. Failing on this task can cause great damage or restrict the robots movement unnecessarily. Due to the security relevance of this problem, great effort is typically spent to design models for individual robots and sensors, and the complexity of such models is correlated to the complexity of the environment and the capabilities of the robot. We present a semi supervised learning approach, where the robot learns its traversability capabilities from a human operating it. From this partially and only positive labeled training data, our approach infers a model for the traversability analysis, thereby requiring very little manual effort for the human. In practical experiments we show that our method can be used for robots that need to reliably navigate on dirt roads as well as for robots that have very restricted traversability capabilities.

Stopping the motion of industrial robots in response to warnings or unexpected sensor data is a special case of trajectory generation. In contrast to emergency stops, here the robot has to satisfy the limits of the acceleration and the jerk. In addition, during the deceleration the robot must follow the path accurately, i.e., the shape of the original path may not be left. This is usually done by scaling the desired velocity. However, for curved paths, e.g. those generated by blending of linear motion commands, by sensor corrections, or directly by splines, this method may leave the desired path. The problem is solved by interpolation using the arc length. In contrast to other methods, here the constraints are considered directly, resulting in a time-efficient computation. Finally, the proposed method prevents a rebound caused by the jerk limits when reaching zero velocity. Experiments are presented using a stiff KUKA robot whose path is exactly tracked during deceleration.

Visual tracking of objects subjected to non-linear motion and appearance changes has shown to be a difficult task in computer vision. While research in visual object tracking has progressed significantly in terms of robust tracking of objects subjected to non-linear motion and appearance changes, these algorithms has shown limited capability for long term tracking of handheld objects during human-object interactions. The failure in tracking is a consequence of abrupt changes in the handheld object motion resulting in tracker drifting off the optimal object space. In this paper, we present a novel 3 layer RGB-D image model formulated with Bayesian filters that tracks handheld object using near constant velocity motion model. Our method divides the image into three layers of abstraction where each encodes visual information of environment, human, object and contributes toward precise localization of the handheld object during tracking. A boundary re-alignment step is introduced during tracking such that the tracker predicted object region is re-aligned to the optimal object region, therefore reducing the likelihood of tracker drifting off the object space. This compensation of the tracker prediction offset enables our algorithm to robustly track handheld object subjected to abrupt changes in motion during manipulation.

In this paper we present a public dataset to evaluate trackers used for human and robot manipulation tasks. For these tasks both high DOF motion and high accuracy is needed. We describe in detail, both the process of recording the sequences and how ground truth data was generated for the videos. The videos are tagged with challenges that a tracker would face while tracking the object. As an initial example, we evaluate the performance of six published trackers [5], [11], [12], [13], [15], [6] and analyse their result. We describe a new evaluation metric to test sensitivity of trackers to speed. A total of 100 annotated and tagged sequences are reported. All the videos, ground truth data, original implementation of trackers and evaluation scripts are made publicly available on the website so others can extend the results on their trackers and evaluation.

This paper presents a method to localize a magnetically guided catheter using a single rotating C-Arm. Because the three dimensional position of the catheter tip, when extracted from a single view, contains large uncertainties, we propose to combine multiple views by rotating the C-Arm and using a Kalman filter. The proposed approach yields good tracking performance for several catheter poses and reduces the maximum estimated standard deviation to 0.3 cm after incorporating six views. This represents a 62 percent reduction in three dimensional position uncertainty.

Social insects have evolved to self-assemble ad-hoc structures from their bodies to quickly adapt to unexpected obstacles and situations. Inspired by these natural systems, we present an autonomous tread-based robot which is capable of using its own body as a building block for assembling structures. We analytically assess the optimality of the robot design, and experimentally test its ability to climb over like robots under varying conditions. Finally, using a simple self-assembly algorithm relying on only local sensing, robot prototypes are used to demonstrate the self-assembly of a 2D pyramid structure.

This paper describes an acoustic navigation system that provides mesoscale coverage (hundreds of kilometers) under the ice and presents results from the first multi-month deployment in the Arctic. The hardware consists of ice-tethered acoustic navigation beacons transmitting at 900 Hz that broadcast their latitude and longitude plus several bytes of optional control data. The real-time under-ice navigation algorithm, based on a Kalman filter, uses time-of-flight measurements from these sources to simultaneously estimate vehicle position and depth-averaged local currents. The algorithm described herein was implemented on Seagliders, a type of autonomous underwater glider (AUG), but the underlying theory is applicable to other autonomous underwater vehicles (AUVs). As part of an extensive field campaign from March to September 2014, eleven acoustic sources and four Seagliders were deployed to monitor the seasonal melt of the marginal ice zone (MIZ) in the Beaufort and northern Chukchi Seas. Beacon-to-beacon performance was excellent due to a sound duct at 100 m depth where the transmitters were positioned; the travel-time error at 200 km has a standard deviation of 40 m when sound-speed is known, and ranges in excess of 400 km were obtained. Results with the Seagliders, which were not regularly within the duct, showed reliable acoustic ranges up to 100 km and more sparse but repeatable range measurements to over 400 km. Navigation results are reported for the real-time algorithm run in post-processing mode, using data from a 295-hour segment with significant time spent under ice.

Space exploratory rovers do well in autonomous or composite semi-autonomous exploration of extraterrestrial surfaces, yet their localization relies on the particular spot they had landed, rather than being universal, i.e. based on the absolute coordinate system of the explored planet. The idea underlaying the work presented in this paper is the transition from the relative to absolute localization by inspecting common Regions of Interest (ROIs) on both rover and orbital imagery. In order to achieve that we propose a method comprising an offline and an onboard procedure. Particularly, prior to the mission the orbital images of the intended landing area are examined to extract ROIs and to construct an offline Global Network (GN). The onboard procedure is based on the rover's self localization which is performed via an inertial aided visual odometry (VO). During its roaming the rover extracts ROIs from the ground and forms a Local Network (LN). The last is iteratively matched with the GN by a specifically designed matching procedure based on Data-Aligned Rigidity-Constrained Exhaustive Search (DARCES). The proposed method is tested on real representative data collected during the ESA Seeker activity. The results indicate that the self-localization of a planetary rover in an absolute frame of reference is feasible, provided that the area includes few discriminative ROIs.

This paper considers pick-and-place tasks using aerial vehicles equipped with manipulators. The main focus is on the development and experimental validation of a nonlinear model-predictive control methodology to exploit the multi-body system dynamics and achieve optimized performance. At the core of the approach lies a sequential Newton method for unconstrained optimal control and a high-frequency low-level controller tracking the generated optimal reference trajectories. A low cost quadrotor prototype with a simple manipulator extending more than twice the radius of the vehicle is designed and integrated with an on-board vision system for object tracking. Experimental results show the effectiveness of model-predictive control to motivate the future use of real-time optimal control in place of standard ad-hoc gain scheduling techniques.

Manipulation planning from high-level task specifications, even though highly desirable, is a challenging problem. The large dimensionality of manipulators and complexity of task specifications make the problem computationally intractable. This work introduces a manipulation planning framework with linear temporal logic (LTL) specifications. The use of LTL as the specification language allows the expression of rich and complex manipulation tasks. The framework deals with the state-explosion problem through a novel abstraction technique. Given a robotic system, a workspace consisting of obstacles, manipulable objects, and locations of interest, and a co-safe LTL specification over the objects and locations, the framework computes a motion plan to achieve the task through a synergistic multi-layered planning architecture. The power of the framework is demonstrated through case studies, in which the planner efficiently computes plans for complex tasks. The case studies also illustrate the ability of the framework in intelligently moving away objects that block desired executions without requiring backtracking.

Life-long visual localization is one of the most challenging topics in robotics over the last few years. The difficulty of this task is in the strong appearance changes that a place suffers due to dynamic elements, illumination, weather or seasons. In this paper, we propose a novel method (ABLE-M) to cope with the main problems of carrying out a robust visual topological localization along time. The novelty of our approach resides in the description of sequences of monocular images as binary codes, which are extracted from a global LDB descriptor and efficiently matched using FLANN for fast nearest neighbor search. Besides, an illumination invariant technique is applied. The usage of the proposed binary description and matching method provides a reduction of memory and computational costs, which is necessary for long-term performance. Our proposal is evaluated in different life-long navigation scenarios, where ABLE-M outperforms some of the main state-of-the-art algorithms, such as WI-SURF, BRIEF-Gist, FAB-MAP or SeqSLAM. Tests are presented for four public datasets where a same route is traversed at different times of day or night, along the months or across all four seasons.

Most manipulation tasks can be decomposed into a sequence of phases, where the robot's actions have different effects in each phase. The robot can perform actions to transition between phases and, thus, alter the effects of its actions, e.g. grasp an object in order to then lift it. The robot can thus reach a phase that affords the desired manipulation. In this paper, we present an approach for exploiting the phase structure of tasks in order to learn manipulation skills more efficiently. Starting with human demonstrations, the robot learns a probabilistic model of the phases and the phase transitions. The robot then employs model-based reinforcement learning to create a library of motor primitives for transitioning between phases. The learned motor primitives generalize to new situations and tasks. Given this library, the robot uses a value function approach to learn a high-level policy for sequencing the motor primitives. The proposed method was successfully evaluated on a real robot performing a bimanual grasping task.

Robotic surgery, especially in the field of neurosurgery, can be tremendously improved with the integration of an excellent imaging modality during the procedure. Shape memory alloy (SMA), a high power density and inexpensive MRI-compatible actuator, is therefore being considered as an appropriate actuator for the robot. However, the low control bandwidth of SMA due to the long cooling time makes it undesirable for commercial use. An efficient and low-cost cooling method using water as a coolant that passes through a flexible tube coiled around the SMA spring is proposed to increase the cooling rate of SMA, thereby improving its actuation frequency. SMA constitutive model and heat transfer model have been developed to simulate theoretical behavior of SMA springs in antagonistic configuration. The maximum bandwidth we achieved was 0.333 Hz for tracking of a sinusoidal trajectory of 3 mm peak-to-peak magnitude. We also demonstrated the capability of our cooling system to control the motion of an SMA spring actuated one-DOF MINIR-II robot prototype.

This paper presents a biologically-inspired real-time balance recovery control strategy that is applied to a lower body exoskeleton with variable physical stiffness actuators at its ankle joints. For this purpose, a torsional spring-loaded flywheel model is presented to encapsulate both approximated angular momentum and variable physical stiffness, which are crucial parameters in describing the postural balance. In particular, the incorporation of physical compliance enables us to provide three main contributions: i) A mathematical formulation is developed to express the relation between the dynamic balance criterion ZMP and the physical ankle joint stiffness. Therefore, balancing control can be interpreted in terms of ankle joint stiffness regulation. ii) ‘Variable physical’ stiffness is utilized in the bipedal robot balance control task for the first time in the literature, to the authors' knowledge. iii) The variable physical stiffness strategy is compared with the optimal constant stiffness strategy by conducting experiments on our exoskeleton robot. The results indicate that the proposed method provides a favorable balancing control performance to cope with unperceived perturbations, in terms of center of mass position regulation, ZMP error and mechanical power.