IROS 2018 Papers

This paper presents a coverage algorithm for multi-robot systems where the robots are equipped with qualitatively different sensing modalities. Unlike previous approaches to the problem of coverage for teams with heterogeneous sensing capabilities, in this paper the robots have access to information about their neighbors' specific sensor modalities. This knowledge affords the ability of ensuring that no robot is tasked with covering features in a region without the required sensing modalities. With this information, a robot can determine which of its neighbors it should coordinate with to cover the environmental features in a region while ignoring robots that are not equipped with that particular sensory capability. We derive a distributed control algorithm that allows the robots to move in a direction of descent relative to a novel locational cost, in order to minimize it. The performance of the algorithm is evaluated on a real robotic platform.

Typically, autonomous cars optimize for a combination of safety, efficiency, and driving quality. But as we get better at this optimization, we start seeing behavior go from too conservative to too aggressive. The car's behavior exposes the incentives we provide in its cost function. In this work, we argue for cars that are not optimizing a purely selfish cost, but also try to be courteous to other interactive drivers. We formalize courtesy as a term in the objective that measures the increase in another driver's cost induced by the autonomous car's behavior. Such a courtesy term enables the robot car to be aware of possible irrationality of the human behavior, and plan accordingly. We analyze the effect of courtesy in a variety of scenarios. We find, for example, that courteous robot cars leave more space when merging in front of a human driver. Moreover, we find that such a courtesy term can help explain real human driver behavior on the NGSIM dataset.

Dynamic systems are a practical alternative to motion planning in executing robot actions. They are of particular interest in Learning from Demonstration, as here we aim to carry out actions in a certain fashion, without a model or in-depth knowledge about the world, which might be difficult to achieve with a planner. Using model-based dynamic systems in task space enables robots to flexibly reproduce demonstrated actions. Nevertheless, when dealing with mobile manipulators, we face the challenge of including the kinematic constraints of the robot in the action models. In this paper we propose to couple robot base and end-effector motions generated by arbitrary dynamical systems modulating the base velocity, while respecting the robots kinematic design. To this end we learn an approximation of the inverse reachability in closed form. In real-world robot experiments we demonstrate that we are able to maintain kinematically feasible trajectories in the presence of obstacles and in configurations differing profoundly from the training scene.

The effectiveness of the robot locomotion can be measured using the cost of transport (CoT) which represents the amount of energy that is needed for traversing from one place to another. Terrains excerpt different mechanical properties when crawled by a multi-legged robot, and thus different values of the CoT. It is therefore desirable to estimate the CoT in advance and plan the robot motion accordingly. However, the CoT might not be known prior the robot deployment, e.g., in extraterrestrial missions; hence, a robot has to learn different terrains as it crawls through the environment incrementally. In this work, we focus on estimating the CoT from visual and geometrical data of the crawled terrain. A thorough analysis of different terrain descriptors within the context of incremental learning is presented to select the best performing approach. We report on the achieved results and experimental verification of the selected approaches with a real hexapod robot crawling over six different terrains.

We focus on autonomously generating robot motion for day to day physical tasks that is expressive of a certain style or emotion. Because we seek generalization across task instances and task types, we propose to capture style via cost functions that the robot can use to augment its nominal task cost and task constraints in a trajectory optimization process. We compare two approaches to representing such cost functions: a weighted linear combination of hand-designed features, and a neural network parameterization operating on raw trajectory input. For each cost type, we learn weights for each style from user feedback. We contrast these approaches to a nominal motion across different tasks and for different styles in a user study, and find that they both perform on par with each other, and significantly outperform the baseline. Each approach has its advantages: featurized costs require learning fewer parameters and can perform better on some styles, but neural network representations do not require expert knowledge to design features and could even learn more complex, nuanced costs than an expert can easily design.

Decentralized receding horizon control (D-RHC) provides a mechanism for coordination in multiagent settings without a centralized command center. However, combining a set of different goals, costs, and constraints to form an efficient optimization objective for D-RHC can be difficult. To allay this problem, we use a meta-learning process - cost adaptation - which generates the optimization objective for D-RHC to solve based on a set of human-generated priors (cost and constraint functions) and an auxiliary heuristic. We use this adaptive D-RHC method for control of mesh-networked swarm agents. This formulation allows a wide range of tasks to be encoded and can account for network delays, heterogeneous capabilities, and increasingly large swarms through the adaptation mechanism. We leverage the Unity3D game engine to build a simulator capable of introducing artificial networking failures and delays in the swarm. Using the simulator we validate our method on an example coordinated exploration task. We demonstrate that cost adaptation allows for more efficient and safer task completion under varying environment conditions and increasingly large swarm sizes. We release our simulator and code to the community for future work.

To deploy a robot in a human-centric environment, it is important that the robot is able to continuously acquire and update object categories while working in the environment. Therefore, autonomous robots must have the ability to continuously execute learning and recognition in a concurrent or interleaved fashion. One of the main challenges in unconstrained human environments is to cope with the effects of context change. This paper presents two main contributions: (i) an approach for evaluating open-ended object category learning and recognition methods in multi-context scenarios; (ii) evaluation of different object category learning and recognition approaches regarding their ability to cope with the effects of context change. Off-line evaluation approaches such as cross-validation do not comply with the simultaneous nature of learning and recognition. A teaching protocol, supporting context change, was therefore designed and used in this work for experimental evaluation. Seven learning and recognition approaches were evaluated and compared using the protocol. The best performance, in terms of number of learned categories, was obtained with a recently proposed local variant of Latent Dirichlet Allocation (LDA), closely followed by a Bag-of-Words (BoW) approach. In terms of adaptability, i.e. coping with context change, the best result was obtained with BoW, immediately followed by the local LDA variant.

This work presents an aerial tool towards the autonomous cooperative coverage and inspection of a large scale 3D infrastructure using multiple Unmanned Aerial Vehicles (UAVs). In the presented approach the UAVs are relying only on their onboard computer and sensory system, deployed for inspection of the 3D structure. In this application each agent covers a different part of the scene autonomously, while avoiding collisions. The autonomous navigation of each platform on the designed path is enabled by the localization system that fuses Ultra Wideband with inertial measurements through an Error- State Kalman Filter. The visual information collected from the aerial team is collaboratively processed to create the 3D model. The performance of the overall setup has been experimentally evaluated in realistic wind turbine inspection experiments, providing dense 3D reconstruction of the inspected structures.

This paper describes a cooperative control approach that combines the use of a powered knee joint orthosis along with Functional Electrical Stimulation (FES) for knee joint flexion-extension movement restoration. A closed-loop adaptive control and an open-loop FES of the quadriceps muscle group are combined together to track a desired knee joint angle trajectory of flexion/extension movements. A nonlinear disturbance observer is used to estimate the torque provided by the subject's muscles through the FES. Simulations and experiments with a healthy subject show the feasibility of the proposed approach. Experiments show the repeatability of motion and the complementarity between the torque provided by the quadriceps muscle through FES and the one delivered by the orthosis actuator to ensure satisfactory tracking of the desired trajectory.

In this paper, we design nonlinear state feedback controllers for discrete-time polynomial dynamical systems via the occupation measure approach. We propose the discrete-time controlled Liouville equation, and use it to formulate the controller synthesis problem as an infinite-dimensional linear programming problem on measures, which is then relaxed as finite-dimensional semidefinite programming problems on moments of measures and their duals on sums-of-squares polynomials. Nonlinear controllers can be extracted from the solutions to the relaxed problems. The advantage of the occupation measure approach is that we solve convex problems instead of generally non-convex problems, and the computational complexity is polynomial in the state and input dimensions, and hence the approach is more scalable. In addition, we show that the approach can be applied to over-approximating the backward reachable set of discrete-time autonomous polynomial systems and the controllable set of discrete-time polynomial systems under known state feedback control laws. We illustrate our approach on several dynamical systems.

Snake-like manipulators can navigate and perform manipulation in confined spaces. Their recent implementations in surgical robots attracted a lot of attentions. These slender manipulators usually possess either a hyper-redundant articulated vertebrate structure or a continuum one. Primary design considerations usually converge to a balance between proper workspace and acceptable stiffness. Efforts have hence been constantly made to achieve higher or adjustable stiffness for a manipulator to widen its applications. This paper presents a simple continuum manipulator design with variable stiffness based on redundantly arranged elastic backbones and continuously constrained bending curvature. The design concepts, kinematics, a preliminary formulation for stiffness adjustment, system construction and experimental characterizations are elaborated. The results showed that the manipulator's stiffness can be increased up to 4.71 times of the value without the curvature constraining rod, indicating the efficacy of the proposed idea.

Projection operators are widely employed in multi-objective robot control. It is an open research question how to achieve continuous transitions between different idempotent projectors which is required for dynamic task priority rearrangement. We formalize projection shaping, providing a solution to deal with rank changes in a smooth fashion. Furthermore, we derive meaningful shaping operators and show that damped least squares is a special case of our general formulation. Finally, we extend the Stack-of-Tasks prioritization scheme for continuous priority rearrangement of single task dimensions. Simulation results validate our approach.

Decision making under uncertainty is a challenging task, especially when dealing with complex robotics scenarios. The Partially Observable Markov Decision Process (POMDP) framework, designed to solve this problem, was subject to much work lately. Most POMDP solvers, however, focus on planning in discrete state, action and/or observations spaces, which does not truly reflect the complexity of most real world problems. This paper addresses the issue by devising a method for solving POMDPs with continuous state, action and observations spaces. The proposed planner, Continuous Belief Tree Search (CBTS), uses Bayesian Optimisation (BO) to dynamically sample promising actions while constructing a belief tree. This dynamic sampling allows for richer action selection than offline action discretisation. CBTS is complemented by a novel trajectory generation technique, relying on the theory of Reproducing Kernel Hilbert Spaces (RKHS), yielding trajectories amenable for robotics applications. The resulting trajectory planner kCBTS outperforms other continuous planners on space modelling and robot parking problems.

The brain-machine interface (BMI) has been reported to offer the potential for controlling the assistive robot for the motor impaired people, using the non-invasively obtained electroencephalogram (EEG) signals. However, the EEG based BMI may not be sufficient and stable to drive the robot moving freely in its 2D or 3D workspace. The robot autonomy may provide assistance for the BMI users with the shared control paradigm. Nevertheless, users suffers from several limitations of the current shared control paradigms applied on BMI, e.g., loss of sense of control, high mental workload due to unintuitive control with the human-robot interface and fixed level of assistance. To overcome these drawbacks, we propose a new control paradigm for the robotic arm reaching task where the robot autonomy is dynamically blended with the gaze-BMI control from a user. In this paradigm, the hybrid gaze-BMI constitutes an intuitive and effective input to continuously control the robotic arm end-effector moving freely in its 2D workspace, with an adjustable speed proportional to the motion intention strength. Furthermore, the adjustable level of assistance by our paradigm allows the system to balance the user's capabilities and feelings of control while compensating for the reaching task's difficulty. The proposed paradigm is verified in the task where a healthy subject utilizes the hybrid gaze-BMI to control the robotic arm end-effector reaching for a target object while avoiding the obstacle in the path. The experimental results demonstrate that the movements with our shared control paradigm are safer, more efficient and less difficult than those without shared control.

Shape changing robot is an approach towards locomotion on uncertain terrain due to its omni-directional features. However, the current locomotion method for such robots rely on discontinuous rolling. We propose a free form locomotion: an omni directional continuous crawling for deformable robots. This method introduce continuous shifting of contact surface similar to amoeba movement. A Mochibot that has thirty two telescopic legs is developed to verify the proposed locomotion method. Through the experiments, we have confirmed that the robot can track smooth paths: straight, smooth, and hand written curves. We also evaluate errors between desired and measured trajectories of the robot.

A robot with contact sensing capability can reduce uncertainty relative to the environment by deliberately moving into contact and matching the resulting contact measurement to different possible states in the world. We present a manipulation planner that finds and sequences these actions by reasoning explicitly about the uncertainty over the robot's state. The planner incrementally constructs a policy that covers all possible contact states during a manipulation and finds contingencies for each of them. In contrast to conformant planners (without contingencies), the planned contingent policies are more robust. We demonstrate this in simulated and real-world manipulation experiments. In contrast to POMDP-based planners, we show that our planner can be directly applied to high-dimensional configuration spaces.

Soft artificial skin sensors that can detect contact forces as well as their locations are attractive in various soft robotics applications. However, soft sensors made of polymer materials have inherent limitations of hysteresis and nonlinearity in response, which makes it highly difficult to implement traditional calibration techniques and yields poor estimation performance. In this paper, we propose intelligent algorithms based on machine learning and logics that can improve the performance of soft sensors. The proposed methods in this paper could be solutions to the aforementioned long-standing problems. They can also be used to simplify the system complexity by reducing the number of signal wires. Three machine learning techniques are discussed in this paper: an artificial neural network (ANN), the k-nearest neighbors (k-NN) algorithm, and a recurrent neural network (RNN). The Preisach model of hysteresis and simple logics were used to support these algorithms. We proved that classifying contact locations on a soft sensor is possible using simple algorithms in real time. Also, force estimation of a single contact was possible using an ANN with the Preisach method. Finally, we successfully estimated forces of multiple contact locations by predicting the outputs of mixed RNN results.

The dangerous work situation in industrial leakage accidents urgently needs a flexible and small robot to help workers perform operations and to protect them from being injured. An aerial manipulator system consisting of a hexa-rotor UAV and a one-DOF manipulator is developed, and is used to press an emergency switch to shut off machinery in an emergency. In practical application, an aerial manipulator usually performs contact operations as the UAV platform is in hover flight. The hovering UAV acting as a spring-mass-damper system is firstly proved. Then, based on the derived spring-mass-damper system model and the impedance control algorithm, the force-sensorless contact force control method is presented. That is, the force is indirectly controlled through controlling the UAV's position error and pitch angle simultaneously. The practical operation experiment of pressing an emergency button shows that the proposed method is able to control the contact force as the aerial manipulator interacts with the external environment.

Grippers made from soft elastomers are able to passively and gently adapt to their targets allowing deformable objects to be grasped safely without causing bruise or damage. However, it is difficult to regulate the contact forces due to the lack of contact feedback for such grippers. In this paper, a modular soft gripper is presented utilizing interchangeable soft pneumatic actuators with embedded flex sensors as fingers of the gripper. The fingers can be assembled in different configurations using 3D printed connectors. The paper investigates the potential of utilizing the simple sensory feedback from the flex and pressure sensors to make additional meaningful inferences regarding the contact state and grasped object size. We study the effect of the grasped object size and contact type on the combined feedback from the embedded flex sensors of opposing fingers. Our results show that a simple linear relationship exists between the grasped object size and the final flex sensor reading at fixed input conditions, despite the variation in object weight and contact type. Additionally, by simply monitoring the time series response from the flex sensor, contact can be detected by comparing the response to the known free-bending response at the same input conditions. Furthermore, by utilizing the measured internal pressure supplied to the soft fingers, it is possible to distinguish between power and pinch grasps, as the contact type affects the rate of change in the flex sensor readings against the internal pressure.