Autonomous Robots最新文献

The accelerated deployment of service robots have spawned a number of algorithm variations to better handle real-world conditions. Many local trajectory planning techniques have been deployed on practical robot systems successfully. While most formulations of Dynamic Window Approach and Model Predictive Control can progress along paths and optimize for additional criteria, the use of pure path tracking algorithms is still commonplace. Decades later, Pure Pursuit and its variants continues to be one of the most commonly utilized classes of local trajectory planners. However, few Pure Pursuit variants have been proposed with schema for variable linear velocities—they either assume a constant velocity or fails to address the point at all. This paper presents a variant of Pure Pursuit designed with additional heuristics to regulate linear velocities, built atop the existing Adaptive variant. The Regulated Pure Pursuit algorithm makes incremental improvements on state of the art by adjusting linear velocities with particular focus on safety in constrained and partially observable spaces commonly negotiated by deployed robots. We present experiments with the Regulated Pure Pursuit algorithm on industrial-grade service robots. We also provide a high-quality reference implementation that is freely included ROS 2 Nav2 framework at https://github.com/ros-planning/navigation2 for fast evaluation.

In order to explore and understand the surrounding environment in an efficient manner, humans have developed a set of space-variant vision mechanisms that allow them to actively attend different locations in the surrounding environment and compensate for memory, neuronal transmission bandwidth and computational limitations in the brain. Similarly, humanoid robots deployed in everyday environments have limited on-board resources, and are faced with increasingly complex tasks that require interaction with objects arranged in many possible spatial configurations. The main goal of this work is to describe and overview biologically inspired, space-variant human visual mechanism benefits, when combined with state-of-the-art algorithms for different visual tasks (e.g. object detection), ranging from low-level hardwired attention vision (i.e. foveal vision) to high-level visual attention mechanisms. We overview the state-of-the-art in biologically plausible space-variant resource-constrained vision architectures, namely for active recognition and localization tasks.

Trajectory planning for multiple robots in shared environments is a challenging problem especially when there is limited communication available or no central entity. In this article, we present Real-time planning using Linear Spatial Separations, or RLSS: a real-time decentralized trajectory planning algorithm for cooperative multi-robot teams in static environments. The algorithm requires relatively few robot capabilities, namely sensing the positions of robots and obstacles without higher-order derivatives and the ability of distinguishing robots from obstacles. There is no communication requirement and the robots’ dynamic limits are taken into account. RLSS generates and solves convex quadratic optimization problems that are kinematically feasible and guarantees collision avoidance if the resulting problems are feasible. We demonstrate the algorithm’s performance in real-time in simulations and on physical robots. We compare RLSS to two state-of-the-art planners and show empirically that RLSS does avoid deadlocks and collisions in forest-like and maze-like environments, significantly improving prior work, which result in collisions and deadlocks in such environments.

Trajectory generation for biped robots is very complex due to the challenge posed by real-world uneven terrain. To address this complexity, this paper proposes a data-driven Gait model that can handle continuously changing conditions. Data-driven approaches are used to incorporate the joint relationships. Therefore, the deep learning methods are employed to develop seven different data-driven models, namely DNN, LSTM, GRU, BiLSTM, BiGRU, LSTM+GRU, and BiLSTM+BiGRU. The dataset used for training the Gait model consists of walking data from 10 able subjects on continuously changing inclines and speeds. The objective function incorporates the standard error from the inter-subject mean trajectory to guide the Gait model to not accurately follow the high variance points in the gait cycle, which helps in providing a smooth and continuous gait cycle. The results show that the proposed Gait models outperform the traditional finite state machine (FSM) and Basis models in terms of mean and maximum error summary statistics. In particular, the LSTM+GRU-based Gait model provides the best performance compared to other data-driven models.

Monitoring and controlling a large-scale spatiotemporal process can be costly and dangerous for human operators, which can delegate the task to mobile robots for improved efficiency at a lower cost. The complex evolution of the spatiotemporal process and limited onboard resources of the robots motivate a holistic design of the robots’ actions to complete the tasks efficiently. This paper describes a cooperative framework for estimating and controlling a spatiotemporal process using a team of mobile robots that have limited onboard resources. We model the spatiotemporal process as a 2D diffusion equation that can characterize the intrinsic dynamics of the process with a partial differential equation (PDE). Measurement and actuation of the diffusion process are performed by mobile robots carrying sensors and actuators. The core of the framework is a nonlinear optimization problem, that simultaneously seeks the actuation and guidance of the robots to control the spatiotemporal process subject to the PDE dynamics. The limited onboard resources are formulated as inequality constraints on the actuation and speed of the robots. Extensive numerical studies analyze and evaluate the proposed framework using nondimensionalization and compare the optimal strategy to baseline strategies. The framework is demonstrated on an outdoor multi-quadrotor testbed using hardware-in-the-loop simulations.

Discharge of mine tailings significantly impacts the ecological status of the sea. Methods to efficiently monitor the extent of dispersion is essential to protect sensitive areas. By combining underwater robotic sampling with ocean models, we can choose informative sampling sites and adaptively change the robot’s path based on in situ measurements to optimally map the tailings distribution near a seafill. This paper creates a stochastic spatio-temporal proxy model of dispersal dynamics using training data from complex numerical models. The proxy model consists of a spatio-temporal Gaussian process model based on an advection–diffusion stochastic partial differential equation. Informative sampling sites are chosen based on predictions from the proxy model using an objective function favoring areas with high uncertainty and high expected tailings concentrations. A simulation study and data from real-life experiments are presented.

We present a biologically inspired design for swarm foraging based on ant’s pheromone deployment, where the swarm is assumed to have very restricted capabilities. The robots do not require global or relative position measurements and the swarm is fully decentralized and needs no infrastructure in place. Additionally, the system only requires one-hop communication over the robot network, we do not make any assumptions about the connectivity of the communication graph and the transmission of information and computation is scalable versus the number of agents. This is done by letting the agents in the swarm act as foragers or as guiding agents (beacons). We present experimental results computed for a swarm of Elisa-3 robots on a simulator, and show how the swarm self-organizes to solve a foraging problem over an unknown environment, converging to trajectories around the shortest path, and test the approach on a real swarm of Elisa-3 robots. At last, we discuss the limitations of such a system and propose how the foraging efficiency can be increased.

This paper propose a hierarchical path planning algorithm that first captures the local crowd movement around the robot using RGB camera combined with LiDAR and predicts the movement of people nearby the robot, and then generates appropriate global path for the robot using the global path planner with the crowd information. After deciding the global path, the low-level control system receives the prediction results of the crowd and high-level global path, and generates the actual speed control commands for the robot after considering the social norms. With the high accuracy of computer vision for human recognition and the high precision of LiDAR, the system is able to accurately track the surrounding human locations. Through high-level path planning, the robot can use different movement strategies in different scenarios, while the crowd prediction allows the robot to generate more efficient and socially acceptable paths. With this system, even in a highly dynamic environment caused by the crowd, the robot can still plan an appropriate path reach the destination without causing psychological discomfort to others successfully.

This paper presents a framework to enable a team of heterogeneous mobile robots to model and sense a multiscale system. We propose a coupled strategy, where robots of one type collect high-fidelity measurements at a slow time scale and robots of another type collect low-fidelity measurements at a fast time scale, for the purpose of fusing measurements together. The multiscale measurements are fused to create a model of a complex, nonlinear spatiotemporal process. The model helps determine optimal sensing locations and predict the evolution of the process. Key contributions are: (i) consolidation of multiple types of data into one cohesive model, (ii) fast determination of optimal sensing locations for mobile robots, and (iii) adaptation of models online for various monitoring scenarios. We illustrate the proposed framework by modeling and predicting the evolution of an artificial plasma cloud. We test our approach using physical marine robots adaptively sampling a process in a water tank.

Robotic cutting of soft materials is critical for applications such as food processing, household automation, and surgical manipulation. As in other areas of robotics, simulators can facilitate controller verification, policy learning, and dataset generation. Moreover, differentiable simulators can enable gradient-based optimization, which is invaluable for calibrating simulation parameters and optimizing controllers. In this work, we present DiSECt: the first differentiable simulator for cutting soft materials. The simulator augments the finite element method with a continuous contact model based on signed distance fields, as well as a continuous damage model that inserts springs on opposite sides of the cutting plane and allows them to weaken until zero stiffness, enabling crack formation. Through various experiments, we evaluate the performance of the simulator. We first show that the simulator can be calibrated to match resultant forces and deformation fields from a state-of-the-art commercial solver and real-world cutting datasets, with generality across cutting velocities and object instances. We then show that Bayesian inference can be performed efficiently by leveraging the differentiability of the simulator, estimating posteriors over hundreds of parameters in a fraction of the time of derivative-free methods. Next, we illustrate that control parameters in the simulation can be optimized to minimize cutting forces via lateral slicing motions. Finally, we conduct experiments on a real robot arm equipped with a slicing knife to infer simulation parameters from force measurements. By optimizing the slicing motion of the knife, we show on fruit cutting scenarios that the average knife force can be reduced by more than (40%) compared to a vertical cutting motion. We publish code and additional materials on our project website at https://diff-cutting-sim.github.io.