Learning From Demonstration Methods Research Articles

Robust Learning from Demonstration Based on GANs and Affine Transformation

Collaborative robots face barriers to widespread adoption due to the complexity of programming them to achieve human-like movement. Learning from demonstration (LfD) has emerged as a crucial solution, allowing robots to learn tasks directly from expert demonstrations, offering versatility and an intuitive programming approach. However, many existing LfD methods encounter issues such as convergence failure and lack of generalization ability. In this paper, we propose: (1) a generative adversarial network (GAN)-based model with multilayer perceptron (MLP) architecture, coupled with a novel loss function designed to mitigate convergence issues; (2) an affine transformation-based generalization method aimed at enhancing LfD tasks by improving their generalization performance; (3) a data preprocessing method tailored to facilitate deployment on robotics platforms. We conduct experiments on a UR5 robotic platform tasked with handwritten digit recognition. Our results demonstrate that our proposed method significantly accelerates generation speed, achieving a remarkable processing time of 23 ms, which is five times faster than movement primitives (MPs), while preserving key features from demonstrations. This leads to outstanding convergence and generalization performance.

Learning-Based Kinematic Control Using Position and Velocity Errors for Robot Trajectory Tracking

In this article, we address the trajectory tracking problem using the learning from demonstration (LFD) method. By using the LFD method, the parameter adjusting problem in the tracking controller is avoided. Consequently, a strategy can be provided to users with limited parameter adjusting experience. The kinematic tracking problem is formulated as a second-order system and the objective is to simultaneously reduce the errors in position and velocity. The extreme learning machines (ELM) algorithm is applied in the controller design. The velocity and position are utilized as the inputs and the output is the robot corrected kinematic movement. The controller parameters are learned from the desired human or programming demonstrations taking into consideration the stability constraints. In this work, we analyze the system local and global asymptotic stability in detail. The effectiveness of the proposed strategy is demonstrated by simulation comparisons and a practical experiment using a KUKA robot manipulator.

Teaching robots to perform quasi-repetitive construction tasks through human demonstration

Robots can assist workers in performing physically-demanding construction tasks, which are typically quasi-repetitive, wherein the geometry of the workspace is dissimilar despite similar tasks. As a result, robots must determine motion trajectories based on the encountered workspace geometry. Learning from Demonstration (LfD) methods have the potential to be used in teaching robots specific tasks through human demonstration, such that robots can then perform learned tasks under different conditions. In this paper, the LfD method is investigated to teach robots how to perform quasi-repetitive construction tasks. Considering ceiling tile installation as the experimental process, the tasks of maneuvering and positioning tiles in a ceiling grid are defined as the target knowledge to be learned. Using a set of human demonstration videos, the designed approach first translates the physical work context, e.g., the pose of the tile, to the target digital twin, i.e., the workspace as-perceived by the robot. The Reinforcement Learning method is then applied to generate the control policy for the robot to perform the subsequent tasks. The proposed method is evaluated in the Robot Operating System (ROS) Gazebo simulator using a KUKA mobile industrial robotic arm emulator and 60 different scenes as test cases. The results show a 78% success rate in installing ceiling tiles based on 3000 virtual and 85 real demonstration videos. The success rate tends to continually rise with an increase in the number of real demonstration videos, confirming the promise and applicability of the LfD method in teaching robot apprentices to perform quasi-repetitive tasks on construction sites.

Robust Learning From Demonstrations With Mixed Qualities Using Leveraged Gaussian Processes

In this paper, we focus on the problem of learning from demonstration (LfD) where demonstrations with different proficiencies are provided without labeling. To this end, we model multiple policies with different qualities as correlated Gaussian processes and present a leverage optimization method that estimates the leverage of each policy where the difference between two leverages defines the correlation between the corresponding policies. To recover a single policy function of an expert, we present a sparsity constraint on the leverage parameters. We first show that the proposed leverage optimization method can recover the correlations between sensory fields where the fields are realized from correlated Gaussian processes and sensor measurements are collected from the fields. Furthermore, we applied the proposed method to autonomous driving experiments, where demonstrations are collected from three different driving modes. While the driving policies are not realized from correlated processes, the proposed method assigns reasonable leverages to the driving demonstrations. The estimated driving policy of an expert, which incorporates the optimized leverages, outperforms previous LfD methods in terms of both safety and driving quality.

Multi-level control architecture for Bionic Handling Assistant robot augmented by learning from demonstration for apple-picking

ABSTRACTThe control of soft continuum robots is challenging owing to their mechanical elasticity and complex dynamics. An additional challenge emerges when we want to apply Learning from Demonstration (LfD) and need to collect necessary demonstrations due to the inherent control difficulty. In this paper, we provide a multi-level architecture from low-level control to high-level motion planning for the Bionic Handling Assistant (BHA) robot. We deploy learning across all levels to enable the application of LfD for a real-world manipulation task. To record the demonstrations, an actively compliant controller is used. A variant of dynamical systems' application that are able to encode both position and orientation then maps the recorded 6D end-effector pose data into a virtual attractor space. A recent LfD method encodes the pose attractors within the same model for point-to-point motion planning. In the proposed architecture, hybrid models that combine an analytical approach and machine learning techniques are used to overcome the inherent slow dynamics and model imprecision of the BHA. The performance and generalization capability of the proposed multi-level approach are evaluated in simulation and with the real BHA robot in an apple-picking scenario which requires high accuracy to control the pose of the robot's end-effector.

A Bayesian Approach to Policy Recognition and State Representation Learning.

Learning from demonstration (LfD) is the process of building behavioral models of a task from demonstrations provided by an expert. These models can be used, e.g., for system control by generalizing the expert demonstrations to previously unencountered situations. Most LfD methods, however, make strong assumptions about the expert behavior, e.g., they assume the existence of a deterministic optimal ground truth policy or require direct monitoring of the expert's controls, which limits their practical use as part of a general system identification framework. In this work, we consider the LfD problem in a more general setting where we allow for arbitrary stochastic expert policies, without reasoning about the optimality of the demonstrations. Following a Bayesian methodology, we model the full posterior distribution of possible expert controllers that explain the provided demonstration data. Moreover, we show that our methodology can be applied in a nonparametric context to infer the complexity of the state representation used by the expert, and to learn task-appropriate partitionings of the system state space.

Guiding Trajectory Optimization by Demonstrated Distributions

Trajectory optimization is an essential tool for motion planning under multiple constraints of robotic manipulators. Optimization-based methods can explicitly optimize a trajectory by leveraging prior knowledge of the system and have been used in various applications such as collision avoidance. However, these methods often require a hand-coded cost function in order to achieve the desired behavior. Specifying such cost function for a complex desired behavior, e.g., disentangling a rope, is a nontrivial task that is often even infeasible. Learning from demonstration (LfD) methods offer an alternative way to program robot motion. LfD methods are less dependent on analytical models and instead learn the behavior of experts implicitly from the demonstrated trajectories. However, the problem of adapting the demonstrations to new situations, e.g., avoiding newly introduced obstacles, has not been fully investigated in the literature. In this letter, we present a motion planning framework that combines the advantages of optimization-based and demonstration-based methods. We learn a distribution of trajectories demonstrated by human experts and use it to guide the trajectory optimization process. The resulting trajectory maintains the demonstrated behaviors, which are essential to performing the task successfully, while adapting the trajectory to avoid obstacles. In simulated experiments and with a real robotic system, we verify that our approach optimizes the trajectory to avoid obstacles and encodes the demonstrated behavior in the resulting trajectory.

Trajectory Generation With Spatio-Temporal Templates Learned From Demonstrations

Learning from demonstration (LfD) represents an effective method for generating a trajectory with prior knowledge obtained from demonstrations. The adaption and generalization abilities are crucial to the application of the LfD method. We present a novel trajectory generation method based on spatio-temporal templates. A trajectory is generated by evaluating the similarity between a spatio-temporal template and features of the generated trajectory. In this way, the spatio-temporal features of the trajectory are maintained based on the prerequisites from task specifications and environmental constraints. The effectiveness of our method is verified through synthetic trajectories and using the Pioneer-3AT wheeled mobile robot platform.