IROS 2015 Papers

This paper presents a practical path-tracking method for a mobile robot with only a downward camera facing the passage plane. A unique algorithm for tracking and searching ground images with natural texture is used to localize the robot without a feature-point extraction scheme commonly used in other visual odometry methods. In our tracking algorithm, groups of reference pixels are used to detect the relative translation and rotation between frames. Furthermore, a reference pixel group of another shape is registered both to record a path and to correct errors accumulated during localization. All image processing and robot control operations are carried out with low memory consumption for image registration and fast calculation times for completing the searches on a laptop PC. We also describe experimental results in which a vehicle developed by the proposed method repeatedly performed precise path tracking under indoor and outdoor environments.

This paper considers the semi-automated robotic surgical procedure for removing the brain tumor margins, where the manual operation is a tedious and time-consuming task for surgeons. We present robust path planning methods for robotic ablation of tumor residues in various shapes, which are represented in point-clouds instead of analytical geometry. Along with the path plans, corresponding metrics are also delivered to the surgeon for selecting the optimal candidate in the automated robotic ablation. The selected path plan is then executed and tested on RAVEN™ II surgical robot platform as part of the semi-automated robotic brain tumor ablation surgery in a simulated tissue phantom.

Multi robot formation is a canonical problem in robotic research. The problem has been examined in neutral environments, where the robots' goal is usually to maintain the formation despite changes in the environment. The problem of multi robot formation has been motivated by natural phenomena such as schools of fish or flocks of birds. While in the natural phenomena the team behavior is responsive to threats, in robotics research of team formation, adversarial presence has been ignored. In this paper we present the problem of adversarial formation, in which a team of robots travels in a connected formation through an adversarial environment that includes threats that may harm the robots. The robots' goal is, therefore, to maximize their chance of traveling through the environment unharmed, where the formation may be used as a mean to achieve this goal. We formally define the problem, present a quantitative measure for evaluating the survivability of the team, and suggest possible solutions to a variant of the problem under certain threat characteristics, optimizing different team survivability criteria. Finally, we discuss the challenges raised by transitioning the discrete representation to a continuous environment in simulation.

In this paper, we solve the path planning problem for a tethered mobile robot, which is connected to a fixed base by a cable of length L. The reachable space of the robot is restricted by the length of the cable and obstacles. The reachable space of the tethered robot can be computed by considering the topology class of the cable. However, it is computationally too expensive to compute this space a-priori. Instead, in this paper, we show how we can plan using a recently-developed variant of A* search, called Multi-Heuristic A*. Normally, the Multi-Heuristic A* algorithm takes in a fixed set of heuristic functions. In our problem, however, the heuristics represent length of paths to the goal along different topology classes, and there can be too many of them and not all the topology classes are useful. To deal with this, we adapt Multi-Heuristic A* to work with a dynamically generated set of heuristic functions. It starts out as a normal weighted A*. Whenever the search gets trapped in a local minimum, we find the proper topology class of the path to escape from it and add the corresponding new heuristic function into the set of heuristic functions considered by the search. We present experimental analysis comparing our approach with weighted A* on planning for a tethered robot in simulation.

This paper presents planning and control of multiple aerial manipulators for cooperative transportation. Individual aerial manipulators which consist of a hexacopter and 2-DOF robotic arm are controlled by an augmented adaptive sliding mode controller based on a closed-chain robot dynamics. The desired path for each aerial manipulator is obtained by using RRT* to transport an object to the desired position. It also considers the constraints about the grasping point at the end effector. To validate the proposed planning and control algorithm, an experimental result with multiple custom-made aerial manipulators is presented, which involves two aerial manipulators tracking the user-guided command and planned trajectory.

Although passive and semi-passive walkers provide energy-efficient locomotion, walking behavior such as a walking cycle cannot be controlled because it depends on a fixed mechanical property such as the center of mass or the foot shape. In this paper, referring to the upper body of a human, we propose a trunk mechanism that has redundant and viscoelastic joints and changes their viscoelasticity with relatively low energy consumption. By using physical and simulation models, we observe the relationship between the viscoelasticity of the trunk joints and the walking cycle. Results of the physical experiment and the simulation show that there are correlations between the joint viscoelasticity and the walking cycle, and thus, the proposed trunk mechanism is considered a suitable mechanism that controls the walking behavior of a semi-passive walker.

This paper explores the use of a single passive design to stabilize frontal plane dynamics for 3D biped walking across a range of forward velocities and/or step lengths. Particular goals are to determine if design of sagittal plane control can be done independently from design of frontal plane stabilization mechanisms, and to explore how dynamic coupling between the two planned motions affects energetic efficiency of walking. Passive dynamic walkers have long utilized curved feet for low energy frontal plane stabilization in 3D walking, with the current design practice of matching the linearized resonance of the curvature to match a particular, steady-state walking gait to achieve stable coupled limit cycle in 3D dynamics. However, practical legged walking systems should operate across a range of velocities and step widths. We examine aspects of the nonlinear dynamics that contribute to the energy efficiency and stability of the system through simulations. Specifically, we focus on the tight coupling between the frontal plane dynamics and stepping speed. We find that roll velocity is strongly coupled to the stepping speed and energy consumption. Our decoupled analysis explains some aspects of the 3D motions; however, the actual effects on cost of transport demonstrate interesting phenomena we had not anticipated. Specifically, while a general trend of increasing cost of transport for 3D vs 2D gaits with stride time does hold in our simulations, the 3D gaits sometimes require less energy than their constrained 2D counterparts, which was a surprising and encouraging result. This work provides a promising direction for the development of practical methods to utilize control designed for planar 2D walking models on more sophisticated 3D dynamic models using little or no additional active control.

In this paper, we discuss the possibility and fundamental properties of passive bipedal walking on a slippery downhill. First, we develop a mathematical model of the passive compass-like biped robot that contacts with the ground incorporating sliding friction which is specified as the Coulomb model. Second, we discuss the possibility of instantaneous stance-leg exchange through mathematical analysis of the collision dynamics. Third, we perform numerical simulations using the model developed, and show that a stable walking gait can be generated with suitable initial conditions and physical parameters. Furthermore, we numerically show that period-doubling bifurcation occurs with the decrease of the coefficient of sliding friction.

This paper deals with the parametric optimum design of a parallel Schönflies-motion robot, named “Ragnar”, designed for fast and flexible pick-and-place applications. The robot architecture admits a rectangular workspace, which can utilize the shop-floor space efficiently. In this work, the parametric models of the transmission quality, elasto-statics and dynamics are established. By taking into consideration of design requirements and pick-and-place trajectory, a comprehensive multi-objective optimization problem is formulated to optimize both kinematic and dynamic performances. The Pareto-front is obtained, which provides optimal solutions to the robot design. Robot prototyping work based on the optimal results is described.

The ubiquitous additive Gaussian noise model is favored in statistical modeling applications for its flexibility and ease of use. Often noise is assumed to be well-represented by a constant covariance, while in reality error characteristics may change predictably. We present an efficient parametric covariance predictor based on the modified Cholesky decomposition that maps from features of the input to covariance matrices. In addition, we discuss fitting the predictor parameters using noise samples with simple regularization techniques. We demonstrate our approach by estimating observation covariances for range-bearing localization with simulated and experimental datasets and show that this results in increased filtering performance compared to traditional covariance adaptation and constant covariance baselines.

Accurate robot-world, hand-eye calibration is crucial to automation tasks. In this paper, we discuss the robot-world, hand-eye calibration problem which has been modeled as the linear relationship AX = ZB, where X and Z are the unknown calibration matrices composed of rotation and translation components. While there are many different approaches to determining X and Z, including linear and iterative methods, we parameterize the rotation components using Euler angles and find a solution using Levenberg-Marquadt iterative approach. We also offer a method to determine A, X, and Z, by formulating the robot-world, hand-eye calibration problem in terms of camera reprojection error. We compare both of these approaches to the state-of-the-art and conclude that our approaches yield lower values of camera reprojection error. In addition, we demonstrate the improved reconstruction accuracy when using the robot-world, hand-eye calibrations produced from our methods.

For satellite servicing, it is necessary to remove a patch of multi-layer insulation (MLI) that covers the access panel. We consider the case where this patch is secured by tape and desire to use ground-based teleoperation to carefully cut the tape on three sides of the patch. Communication delays of several seconds motivate the development of an online method to enable failure detection by the remote (on-orbit) robot system, so that cutting can be stopped without having to wait several seconds for the ground-based operator to observe the failure. This method is based on a model that predicts the force in the direction of cutting. The model parameters are provided by a recursive least squares estimator, with vector-like forgetting factors, that also includes a throttling mechanism to ensure that the estimator is used only when operating conditions and measurements enable reasonable outcomes. During cutting, the predicted force is compared to the measured force to detect various types of failures. Experiments are conducted on a ground-based platform to demonstrate that the proposed estimation system can reliably detect these failures.

Appropriate robot behavior in public, open spaces cannot occur without the ability to automatically detect conversational groups of free-standing people. To this end, we propose an alternating optimization procedure that estimates lower body orientations and detects groups of interacting people. The first task is achieved by tracking the direction of the lower body of the people in the scene based on their position, their head orientation, the location of objects of interest in their vicinity, and their groups. For the second task, we propose a new group detection algorithm based on F-formation detection. This method can reason about lower body orientation distributions, and generates soft group assignments for the orientation trackers. We evaluate the proposed approach on a publicly available dataset, and show that it can improve state-of-the-art detection of non-interacting people without sacrificing group detection accuracy. This is particularly useful for robots since it provides more opportunities for starting interactions and can help estimate disengagement.

Presented is PYROBOTS, a new open-source software toolkit for the executive control of complex robotic systems. Borrowing ideas from previous tools like URBI [1], it proposes a lightweight Python framework to develop (preemptive) concurrent and event-based executive controllers. Designed in a bottom-up fashion, out of the need for a practical, unobstrusive tool suitable for rapid prototyping of complex interaction scenarios, PYROBOTS also exposes several simple yet convenient abstractions for physical resources management and pose representation. While middleware-agnostic, it features specific integration with the ROS middleware. Experimental deployments and stress-tests on several robotic platforms (including PR2 and Nao) in dynamic human environments are reported.

Navigation in unknown, chaotic environments continues to present a significant challenge for the robotics community. Lighting changes, self-similar textures, motion blur, and moving objects are all considerable stumbling blocks for state-of-the-art vision-based navigation algorithms. In this paper we present a novel technique for improving localization accuracy within a visual-inertial navigation system (VINS). We make use of training data to learn a model for the quality of visual features with respect to localization error in a given environment. This model maps each visual observation from a predefined prediction space of visual-inertial predictors onto a scalar weight, which is then used to scale the observation covariance matrix. In this way, our model can adjust the influence of each observation according to its quality. We discuss our choice of predictors and report substantial reductions in localization error on 4 km of data from the KITTI dataset, as well as on experimental datasets consisting of 700 m of indoor and outdoor driving on a small ground rover equipped with a Skybotix VI-Sensor.

The problem of uniquely identifying an entity described in natural language, known as reference resolution, has become recognized as a critical problem for the field of robotics, as it is necessary in order for robots to be able to discuss, reason about, or perform actions involving any people, locations, or objects in their environments. However, most existing algorithms for reference resolution are domain-specific and limited to environments assumed to be known a priori. In this paper we present an algorithm for reference resolution which is both domain independent and designed to operate in an open world. We call this algorithm POWER: Probabilistic Open-World Entity Resolution. We then present the results of an empirical study demonstrating the success of POWER both in properly identifying the referents of referential expressions and in properly modifying the world model based on such expressions.

Disaster response is one of the most critical social issues and introduces quite a few research themes for the AI planning area. Robocup Rescue provides a platform to simulate the rescue process in a city when an earthquake happens. Existing methods consist of multi-agent methods that use greedy heuristics. These methods scale to large maps but suffer from volatile performance under different scenarios. In this work, we propose a planning framework to boost the performance on Robocup Rescue given several policies from the competition to be used as components. More specifically, we use an online POMDP algorithm with macro-actions and restrict it to plan within the space of tasks performed by the agents in the component policies at each time instance. Since the action space contains macro-actions of the component policies, the method is guaranteed to perform at least as well as the best component policy, and possibly better, if sufficient computation is provided. On the other hand, the restriction of the tasks to those suggested by component policies reduces the computational complexity of planning and allows the planning method to be practically applied. Experiment results show that our planner generates better performance than the best component policy for some scenarios and gives performance comparable to the best component policy for the rest.

Efficient object manipulation based only on force feedback typically requires a plan of actively contact-seeking actions to reduce uncertainty over the true environmental model. In principle, that problem could be formulated as a full partially observable Markov decision process (POMDP) whose observations are sensed forces indicating the presence/absence of contacts with objects. Such a naive application leads to a very large POMDP with high-dimensional continuous state, action and observation spaces. Solving such large POMDPs is practically prohibitive. In other words, we are facing three challenging problems: 1) uncertainty over discontinuous contacts with objects; 2) high-dimensional continuous spaces; 3) optimization for not only trajectory cost but also execution time. As trajectory optimization is a powerful model-based method for motion generation, it can handle the last two issues effectively by computing locally optimal trajectories. This paper aims to integrate advantages of trajectory optimization into existing POMDP solvers. The full POMDP formulation is solved using sample-based approaches, where each sampled model is quickly evaluated via trajectory optimization instead of simulating a large number of rollouts. To further accelerate the solver, we propose to integrate temporal abstraction, i.e. macro actions or temporal actions, into the POMDP model. We demonstrate the proposed method on a simulated 7 DoF KUKA arm and a physical Willow Garage PR2 platform. The results show that our proposed method could effectively seek contacts in complex scenarios, and achieve near-optimal performance of path planing.

Mobile humanoid robots have the capability of accomplishing complex tasks in human dominated environments. In order to execute a task with its arm, a robot needs to place its base reasonably first. In previous work, a positionbased reachability map solved this problem by matching a precomputed inverse kinematics database with discretized versions of the task path. However, its capability is restricted to the end effector with which the database was computed. This paper proposes an orientation-based reachability map which supports on-line end effector frame extensions. The added extension capability enables robots to handle partially constrained task paths with different tool frames, which would be non-trivial for previous methods without recomputing a reachability map. We discuss the differences between the new method and its predecessors, and analyze extension transformation matrix to provide mathematical proof for the extension capability of our reachability map. Our experimental results show that the orientation-based reachability map is as fast as the positionbased reachability map while benefiting from the additional extension capability.