Rapidly random-exploring tree (RRT) and its variants are very popular due to their ability to quickly and efficiently explore the state space. However, they suffer sensitivity to the initial solution and slow convergence to the optimal solution, which means that they consume a lot of memory and time to find the optimal path. It is critical to quickly find a short path in many applications such as the autonomous vehicle with limited power/fuel. To overcome these limitations, we propose a novel optimal path planning algorithm based on the convolutional neural network (CNN), namely the neural RRT* (NRRT*). The NRRT* utilizes a nonuniform sampling distribution generated from a CNN model. The model is trained using quantities of successful path planning cases. In this article, we use the A* algorithm to generate the training data set consisting of the map information and the optimal path. For a given task, the proposed CNN model can predict the probability distribution of the optimal path on the map, which is used to guide the sampling process. The time cost and memory usage of the planned path are selected as the metric to demonstrate the effectiveness and efficiency of the NRRT*. The simulation results reveal that the NRRT* can achieve convincing performance compared with the state-of-the-art path planning algorithms.

Ultrasound (US) acquisition in standard clinical procedures requires manual operation of the probe based on the interpretation of the image. A robotic system for autonomous US acquisitions holds great promise to relieve the workload of sonographers, improve access to care, yield more standardized imaging results, and circumvent the need for direct patient contact. In this paper, we provide an overview of the current status of systems and techniques for autonomous robotic US acquisitions in the literature. Specifically, we first present a framework to standardize the increasing levels of autonomy in robotic ultrasonography and briefly discuss the non-autonomous and semi-autonomous systems. Then, we systematically review the existing autonomous robotic systems developed for extracorporeal US acquisitions based on the hardware components, workflow and applications. Besides, we summarize three important topics related with the autonomous US acquisitions, including the scanning path planning, contact force control and image quality optimization, and introduce current methods and techniques proposed on each topic. Finally, we discuss the challenges and provide some potential directions for future research. We believe that the presented efforts in autonomous robotic US acquisitions will pave the way for a promising future of medical care and benefit a broad spectrum of clinical procedures.

Recently Koopman operator has become a promising data-driven tool to facilitate real-time control for unknown nonlinear systems. It maps nonlinear systems into equivalent linear systems in embedding space, ready for real-time linear control methods. However, designing an appropriate Koopman embedding function remains a challenging task. Furthermore, most Koopman-based algorithms only consider nonlinear systems with linear control input, resulting in lousy prediction and control performance when the system is fully nonlinear with the control input. In this work, we propose an end-to-end deep learning framework to learn the Koopman embedding function and Koopman Operator together to alleviate such difficulties. We first parameterize the embedding function and Koopman Operator with the neural network and train them end-to-end with the K-steps loss function. Then, an auxiliary control network is augmented to encode the nonlinear state-dependent control term to model the nonlinearity in the control input. This encoded term is considered the new control variable instead to ensure linearity of the modeled system in the embedding system. We next deploy Linear Quadratic Regulator (LQR) on the linear embedding space to derive the optimal control policy and decode the actual control input from the control net. Experimental results demonstrate that our approach outperforms other existing methods, reducing the prediction error by order of magnitude and achieving superior control performance in several nonlinear dynamic systems like damping pendulum, CartPole, and the seven DOF robotic manipulator.

In a human-robot coexisting environment, it is pivotal for a mobile service robot to arrive at the goal position safely and efficiently. In this article, an elastic band-based rapidly exploring random tree (EB-RRT) algorithm is proposed to achieve real-time optimal motion planning for the mobile robot in the dynamic environment, which can maintain a homotopy optimal trajectory based on current heuristic trajectory. Inspired by the EB method, we propose a hierarchical framework consisting of two planners. In the global planner, a time-based RRT algorithm is used to generate a feasible heuristic trajectory for a specific task in the dynamic environment. However, this heuristic trajectory is nonoptimal. In the dynamic replanner, the time-based nodes on the heuristic trajectory are updated due to the internal contraction force and the repulsive force from the obstacles. In this way, the heuristic trajectory is optimized continuously, and the final trajectory can be proved to be optimal in the homotopy class of the heuristic trajectory. Simulation experiments reveal that compared with two stateof-the-art algorithms, our proposed method can achieve better performance in dynamic environments.

Mobile robot autonomous path planning is an essential factor for its wide deployment in real-world applications. Conventional sampling-based algorithms have gained tremendous success in the path planning field, but they usually take much time to find the optimal solution so that the planning quality (evaluated with time cost and path length) cannot be guaranteed. In this paper, based on Gaussian Mixture Regression (GMR) and the family of Rapidly-exploring Random Tree (RRT) schemes, we propose the GMR-RRT* algorithm to achieve fast path planning for mobile robots. The proposed GMR-RRT* consists of learning navigation behaviors from human demonstrations and planning a high-quality path for the robot. Using the GMR, the key features of human demonstrations are captured to form a probability density distribution of the human trajectory in the current environment. This distribution is further utilized to guide the RRT scheme’s sampling process to generate a feasible path in the current environment quickly. We test the proposed GMR-RRT* in different environments, comparing it with three state-of-the-art sampling-based algorithms. The experimental results demonstrate that the GMR-RRT* algorithm can achieve better performance in terms of time cost, memory usage, and path length.

A fundamental task in robotics is to plan collision‐free motions among a set of obstacles. Recently, learning‐based motion‐planning methods have shown significant advantages in solving different planning problems in high‐dimensional spaces and complex environments. This article serves as a survey of various different learning‐based methods that have been applied to robot motion‐planning problems, including supervised, unsupervised learning, and reinforcement learning. These learning‐based methods either rely on a human‐crafted reward function for specific tasks or learn from successful planning experiences. The classical definition and learning‐related definition of motion‐planning problem are provided in this article. Different learning‐based motion‐planning algorithms are introduced, and the combination of classical motion‐planning and learning techniques is discussed in detail.