This letter explores the problem of delivering unwieldy objects using nonholonomic mobile bases. We propose a new approach called free pushing to address this challenge. Unlike previous stable pushing methods which maintain a stiff robot-object contact, our approach allows the robot to maneuver around the object while pushing it. It aims to execute continuous pushes without losing contact for improved pushing maneuverability. Additionally, to ensure the feasibility of the planned pushes, a robot-object contact model is developed to account for the shape and kinematics of the robot in pushing modeling and planning. A Model Predictive Controller solves the pushing planning problem in real time. Experimental results show that the proposed method achieves an average success rate of 83% with an accuracy of 0.085 m when pushing to the selected goals. Compared to the baselines, this approach improves the agility and efficiency of mobile pushers. Furthermore, it is robust in achieving the task while tolerating modeling errors.

@en

This letter addresses the problem of pushing manipulation with nonholonomic mobile robots. Pushing is a fundamental skill that enables robots to move unwieldy objects that cannot be grasped. We propose a stable pushing method that maintains stiff contact between the robot and the object to avoid consuming repositioning actions. We prove that a line contact, rather than a single point contact, is necessary for nonholonomic robots to achieve stable pushing. We also show that the stable pushing constraint and the nonholonomic constraint of the robot can be simplified as a concise linear motion constraint. Then the pushing planning problem can be formulated as a constrained optimization problem using nonlinear model predictive control (NMPC). According to the experiments, our NMPC-based planner outperforms a reactive pushing strategy in terms of efficiency, reducing the robot's traveled distance by 23.8% and time by 77.4%. Furthermore, our method requires four fewer hyperparameters and decision variables than the Linear Time-Varying (LTV) MPC approach, making it easier to implement. Real-world experiments are carried out to validate the proposed method with two differential-drive robots, Husky and Boxer, under different friction conditions.

@en

Optimization fabrics are a geometric approach to real-time local motion generation, where motions are designed by the composition of several differential equations that exhibit a desired motion behavior. We generalize this framework to dynamic scenarios and nonholonomic robots and prove that fundamental properties can be conserved. We show that convergence to desired trajectories and avoidance of moving obstacles can be guaranteed using simple construction rules of the components. In addition, we present the first quantitative comparisons between optimization fabrics and model predictive control and show that optimization fabrics can generate similar trajectories with better scalability, and thus, much higher replanning frequency (up to 500 Hz with a 7 degrees of freedom robotic arm). Finally, we present empirical results on several robots, including a nonholonomic mobile manipulator with 10 degrees of freedom and avoidance of a moving human, supporting the theoretical findings.

@en

In this article, we propose a hybrid combination of active inference and behavior trees (BTs) for reactive action planning and execution in dynamic environments, showing how robotic tasks can be formulated as a free-energy minimization problem. The proposed approach allows handling partially observable initial states and improves the robustness of classical BTs against unexpected contingencies while at the same time reducing the number of nodes in a tree. In this work, we specify the nominal behavior offline, through BTs. However, in contrast to previous approaches, we introduce a new type of leaf node to specify the desired state to be achieved rather than an action to execute. The decision of which action to execute to reach the desired state is performed online through active inference. This results in continual online planning and hierarchical deliberation. By doing so, an agent can follow a predefined offline plan while still keeping the ability to locally adapt and take autonomous decisions at runtime, respecting safety constraints. We provide proof of convergence and robustness analysis, and we validate our method in two different mobile manipulators performing similar tasks, both in a simulated and real retail environment. The results showed improved runtime adaptability with a fraction of the hand-coded nodes compared to classical BTs.

@en

The free energy principle (FEP) from neuroscience provides a framework called active inference for the joint estimation and control of state space systems, subjected to colored noise. However, the active inference community has been challenged with the critical task of manually tuning the noise smoothness parameter. To solve this problem, we introduce a novel online noise smoothness estimator based on the idea of free energy principle. We mathematically show that our estimator can converge to the free energy optimum during smoothness estimation. Using this formulation, we introduce a joint state and noise smoothness observer design called DEMs. Through rigorous simulations, we show that DEMs outperforms state-of-the-art state observers with least state estimation error. Finally, we provide a proof of concept for DEMs by applying it on a real life robotics problem - state estimation of a quadrotor hovering in wind, demonstrating its practical use.@en

The free energy principle from neuroscience provides an efficient data-driven framework called the Dynamic Expectation Maximization (DEM), to learn the generative model in the environment. DEM’s growing potential to be the brain-inspired learning algorithm for robots demands a mathematically rigorous analysis using the standard control system tools. Therefore, this paper derives the mathematical proof of convergence for its parameter estimator for linear state space systems, subjected to colored noise. We show that the free energy based parameter learning converges to a stable solution for linear systems. The paper concludes by providing a proof of concept through simulation for a wide range of spring damper systems.

@en

The free energy principle from neuroscience provides a brain-inspired perception scheme through a data-driven model learning algorithm called Dynamic Expectation Maximization (DEM). This paper aims at introducing an exper-imental design to provide the first experimental confirmation of the usefulness of DEM as a state and input estimator for real robots. Through a series of quadcopter flight experiments under unmodelled wind dynamics, we prove that DEM can leverage the information from colored noise for accurate state and input estimation through the use of generalized coordinates. We demonstrate the superior performance of DEM for state es-timation under colored noise with respect to other benchmarks like State Augmentation, SMIKF and Kalman Filtering through its minimal estimation error. We demonstrate the similarities in the performance of DEM and Unknown Input Observer (UIO) for input estimation. The paper concludes by showing the influence of prior beliefs in shaping the accuracy-complexity trade-off during DEM's estimation.

@en

The free energy principle from neuroscience has recently gained traction as one of the most prominent brain theories that can emulate the brain’s perception and action in a bio-inspired manner. This renders the theory with the potential to hold the key for general artificial intelligence. Leveraging this potential, this paper aims to bridge the gap between neuroscience and robotics by reformulating an FEP-based inference scheme—Dynamic Expectation Maximization—into an algorithm that can perform simultaneous state, input, parameter, and noise hyperparameter estimation of any stable linear state space system subjected to colored noises. The resulting estimator was proved to be of the form of an augmented coupled linear estimator. Using this mathematical formulation, we proved that the estimation steps have theoretical guarantees of convergence. The algorithm was rigorously tested in simulation on a wide variety of linear systems with colored noises. The paper concludes by demonstrating the superior performance of DEM for parameter estimation under colored noise in simulation, when compared to the state-of-the-art estimators like Sub Space method, Prediction Error Minimization (PEM), and Expectation Maximization (EM) algorithm. These results contribute to the applicability of DEM as a robust learning algorithm for safe robotic applications.@en

We present a fault tolerant control scheme for robot manipulators based on active inference. The proposed solution makes use of the sensory prediction errors in the free-energy to simplify the residuals and thresholds generation for fault detection and isolation and does not require additional controllers for fault recovery. Results validating the benefits in a simulated 2DOF manipulator are presented and the limitations of the current approach are highlighted.

@en

The free energy principle from neuroscience provides a biologically plausible solution to the brain's inference mechanism. This paper reformulates this theory to design a brain-inspired state and input estimator for a linear time-invariant state space system with colored noise. This reformulation for linear systems bridges the gap between the neuroscientific theory and control theory, therefore opening up the possibility of evaluating it under the hood of standard control approaches. Through rigorous simulations under colored noises, the observer is shown to outperform Kalman Filter and Unknown Input Observer with minimal error in state and input estimation. It is tested against a wide range of scenarios and the proof of concept is demonstrated by applying it on a real system.

@en

The main task of robotic grippers, holding an object, does not require work theoretically. Yet grippers consume significant amounts of energy in practice. This paper presents an approach for designing an energy-saving drive for robotic grippers employing a Statically Balanced Force Amplifier (SBFA) and a Non-backdrivable mechanism (NBDM). A novel metric (Grip Performance Metric) to systematically evaluate drives regarding their energy consumption, is used in the design phase; afterwards, the realization and testing of a prototype (REED, Robotic Energy-Efficient Drive) are presented. Results show that the actuation force can be reduced by 92%, resulting in energy-savings of 86% for an example task. This shows the potential of drives based on SBFAs and NBDMs to achieve energy-neutral grippers.

@en

In electrically actuated robots most energy losses are due to the heating of the actuators. This energy loss can be greatly reduced with parallel elastic actuators, by optimizing the elastic element such that it delivers most of the required torques. Previously used optimization methods relied on parameterizing the spring characteristic, thereby limiting the set of spring characteristics optimized over and with that the loss reduction that can be obtained. This letter shows that such parametrization is not necessary; a method is presented to compute the optimal characteristic as an analytic function of the trajectory. The efficacy of this method is demonstrated using two examples. The first example considers the optimal spring characteristic for a parallel elastic actuator supporting the human ankle during walking. The second example applies the method in combination with trajectory optimization on a single degree of freedom robot performing a specific pick-and-place task. The task at hand has a height difference between the pick and the place location. With the analytical optimal spring, it is shown that the robot can recover enough of the energy released by the package to function without external electric energy supply.

@en

This article describes Team Delft's robot winning the Amazon Robotics Challenge 2016. The competition involves automating pick and place operations in semi-structured environments, specifically the shelves in an Amazon warehouse.
Team Delft's entry demonstrated that current robot technology can already address most of the challenges in product handling: object recognition, grasping, motion, or task planning; under broad yet bounded conditions. The system combines an industrial robot arm, 3D cameras and a custom gripper. The robot's software is based on the Robot Operating System to implement solutions based on deep learning and other state-of-the-art artificial intelligence techniques, and to integrate them with off-the-shelf components.
From the experience developing the robotic system it was concluded that: 1) the specific task conditions should guide the selection of the solution for each capability required, 2) understanding the characteristics of the individual solutions and the assumptions they embed is critical to integrate a performing system from them, and 3) this characterization can be based on `levels of robot automation'. This paper proposes automation levels based on the usage of information at
design or runtime to drive the robot's behaviour, and uses them to discuss Team Delft's design solution and the lessons learned from this robot development experience.
@en

In this paper, a novel active vision strategy is proposed for optimizing the viewpoint of a robot's vision sensor for a given success criterion. The strategy is based on extremum seeking control (ESC), which introduces two main advantages: 1) Our approach is model free: It does not require an explicit objective function or any other task model to calculate the gradient direction for viewpoint optimization. This brings new possibilities for the use of active vision in unstructured environments, since a priori knowledge of the surroundings and the target objects is not required. 2) ESC conducts continuous optimization backed up with mechanisms to escape from local maxima. This enables an efficient execution of an active vision task. We demonstrate our approach with two applications in the object recognition and manipulation fields, where the model-free approach brings various benefits: for object recognition, our framework removes the dependence on offline training data for viewpoint optimization, and provides robustness of the system to occlusions and changing lighting conditions. In object manipulation, the model-free approach allows us to increase the success rate of a grasp synthesis algorithm without the need of an object model; the algorithm only uses continuous measurements of the objective value, i.e., the grasp quality. Our experiments show that continuous viewpoint optimization can efficiently increase the data quality for the underlying algorithm, while maintaining the robustness.

@en

Increasing grasping efficiency is very important for the robots to grasp unknown objects especially subjected to unfamiliar environments. To achieve this, a new algorithm is proposed based on the C-shape configuration. Specifically, the geometric model of the used under-actuated gripper is approximated as a C-shape. To obtain an appropriate graspable position, this C-shape configuration is applied to fit geometric model of an unknown object. The geometric model of unknown object is constructed by using a single-view partial point cloud. To examine the algorithm using simulations, a comparison of the commonly used motion planners is made. The motion planner with the highest number of solved runs, lowest computing time and the shortest path length is chosen to execute grasps found by this grasping algorithm. The simulation results demonstrate that excellent grasping efficiency is achieved by adopting our algorithm. To validate this algorithm, experiment tests are carried out using a UR5 robot arm and an under-actuated gripper. The experimental results show that steady grasping actions are obtained. Hence, this research provides a novel algorithm for fast grasping of unknown objects.@en

Sampling-based kinodynamic planners, such as Rapidly-exploring Random Trees (RRTs), pose two fundamental challenges: computing a reliable (pseudo-)metric for the distance between two randomly sampled nodes, and computing a steering input to connect the nodes. The core of these challenges is a Two Point Boundary Value Problem, which is known to be NP-hard. Recently, the distance metric has been approximated using supervised learning, reducing computation time drastically. The previous work on such learning RRTs use direct optimal control to generate the data for supervised learning. This paper proposes to use indirect optimal control instead, because it provides two benefits: it reduces the computational effort to generate the data, and it provides a low dimensional parametrization of the action space. The latter allows us to learn both the distance metric and the steering input to connect two nodes. This eliminates the need for a local planner in learning RRTs. Experimental results on a pendulum swing up show 10-fold speed-up in both the offline data generation and the online planning time, leading to at least a 10-fold speed-up in the overall planning time.

@en

Grasp synthesis for unknown objects is a challenging problem as the algorithms are expected to cope with missing object shape information. This missing information is a function of the vision sensor viewpoint. The majority of the grasp synthesis algorithms in literature synthesize a grasp by using one single image of the target object and making assumptions on the missing shape information. On the contrary, this paper proposes the use of robot's depth sensor actively: we propose an active vision methodology that optimizes the viewpoint of the sensor for increasing the quality of the synthesized grasp over time. By this way, we aim to relax the assumptions on the sensor's viewpoint and boost the success rates of the grasp synthesis algorithms. A reinforcement learning technique is employed to obtain a viewpoint optimization policy, and a training process and automated training data generation procedure are presented. The methodology is applied to a simple force-moment balance-based grasp synthesis algorithm, and a thousand simulations with five objects are conducted with random initial poses in which the grasp synthesis algorithm was not able to obtain a good grasp with the initial viewpoint. In 94% of these cases, the policy achieved to find a successful grasp.@en

This paper identifies high performing motion planners among three manipulators when carrying out grasp executions. Simultaneously, this paper presents useful benchmarking data. Sampling-based motion planners of OMPL available for use in MoveIt! are compared by performing several grasping-related motion planning problems. The performance of the planners is measured by means of solved runs, computing time and path length. Based on the results, recommendations are made for planner choice that shows high performance for the used manipulators.@en

This paper identifies the class of actuators called clutched elastic actuators (CEAs). CEAs use clutches to control the energy flow into springs. CEAs in exoskeletons, prostheses, legged robots, and robotic arms have shown the ability to reduce the energy consumption and motor requirements, such as peak torque and peak power. Because of those abilities, they are increasingly used in robotics. In this paper, we categorize existing CEA designs, identify trends in those designs, and provide a method to analyze their functionality. Based on a literature survey, current CEA designs are placed in nine categories, depending on their morphology. The main trend is that CEA designs are becoming more complex, meaning that the number of clutches and springs increases. We show with the introduced mathematical analysis that the functionality can be analyzed with a constraint matrix, a stiffness matrix, and multiplication of a clutch-dependent diagonal matrix with an oriented incidence matrix. This method eases the analysis of the functionality of CEAs. Furthermore, it can lead to new CEA designs in which the number of resulting stiffnesses grows exponentially with the number of springs and clutches.@en