Autonomous Robots最新文献

Collision avoidance is one of the most important topics in the robotics field. In this problem, the goal is to move the robots from initial locations to target locations such that they follow the shortest non-colliding paths in the shortest time and with the least amount of energy. Robot navigation among pedestrians is an example application of this problem which is the focus of this paper. This paper presents a distributed and real-time algorithm for solving collision avoidance problems in dense and complex 2D and 3D environments. This algorithm uses angular calculations to select the optimal direction for the movement of each robot and it has been shown that these separate calculations lead to a form of cooperative behavior among agents. We evaluated the proposed approach on various simulation and experimental scenarios and compared the results with ORCA one of the most important algorithms in this field. The results show that the proposed approach is at least 25% faster than ORCA while is also more reliable. The proposed method is shown to enable fully autonomous navigation of a swarm of Crazyflies.

Dynamic environments challenge existing robot navigation methods, and motivate either stringent assumptions on workspace variation or relinquishing of collision avoidance and convergence guarantees. This paper shows that the latter can be preserved even in the absence of knowledge of how the environment evolves, through a navigation function methodology applicable to sphere-worlds with moving obstacles and robot destinations. Assuming bounds on speeds of robot destination and obstacles, and sufficiently higher maximum robot speed, the navigation function gradient can be used produce robot feedback laws that guarantee obstacle avoidance, and theoretical guarantees of bounded tracking errors and asymptotic convergence to the target when the latter eventually stops moving. The efficacy of the gradient-based feedback controller derived from the new navigation function construction is demonstrated both in numerical simulations as well as experimentally.

Correspondence identification is essential for multi-robot collaborative perception, which aims to identify the same objects in order to ensure consistent references of the objects by a group of robots/agents in their own fields of view. Although recent deep learning methods have shown encouraging performance on correspondence identification, they suffer from two shortcomings, including the inability to address non-covisibility and the inability to quantify and reduce uncertainty to improve correspondence identification. To address both issues, we propose a novel uncertainty-aware deep graph matching method for correspondence identification in collaborative perception. Our new approach formulates correspondence identification as a deep graph matching problem, which identifies correspondences based on deep graph neural network-based features and explicitly quantify uncertainties in the identified correspondences under the Bayesian framework. In addition, we design a novel loss function that explicitly reduces correspondence uncertainty and perceptual non-covisibility during learning. Finally, we design a novel multi-robot sensor fusion method that integrates the multi-robot observations given the identified correspondences to perform collaborative object localization. We evaluate our approach in the robotics applications of collaborative assembly, multi-robot coordination and connected autonomous driving using high-fidelity simulations and physical robots. Experiments have shown that, our approach achieves the state-of-the-art performance of correspondence identification. Furthermore, the identified correspondences of objects can be well integrated into multi-robot collaboration for object localization.

In order to provide adaptive and user-friendly solutions to robotic manipulation, it is important that the agent can learn to accomplish tasks even if they are only provided with very sparse instruction signals. To address the issues reinforcement learning algorithms face when task rewards are sparse, this paper proposes an intrinsic motivation approach that can be easily integrated into any standard reinforcement learning algorithm and can allow robotic manipulators to learn useful manipulation skills with only sparse extrinsic rewards. Through integrating and balancing empowerment and curiosity, this approach shows superior performance compared to other state-of-the-art intrinsic exploration approaches during extensive empirical testing. When combined with other strategies for tackling the exploration challenge, e.g. curriculum learning, our approach is able to further improve the exploration efficiency and task success rate. Qualitative analysis also shows that when combined with diversity-driven intrinsic motivations, this approach can help manipulators learn a set of diverse skills which could potentially be applied to other more complicated manipulation tasks and accelerate their learning process.

Load transportation by Unmanned Aerial Vehicles is a research topic of great interest to the robotic community for its numerous applications in both the civilian and military fields. Attaching a cargo through an elastic cable to a small underactuated UAV such as a quadcopter, which is inherently an unstable system, increases its instability and its underactuated degrees of freedom by three. Moreover, the presence of imperfections in the system such as having the quadcopter’s center of gravity and the cable hanging point arbitrarily shifted from the quadcopter’s geometric centroid further complicates the system. In this paper, a new nonlinear nine degree-of-freedom mathematical model is formulated for a quadcopter when its center of gravity is shifted from its geometric centroid and when a cable-suspended load is attached at an arbitrary position. Thus, a novel adaptive controller based on Linear Quadratic Regulator is designed to control the position and attitude of the quadcopter while minimizing the swinging and radial motions of the suspended load. Subsequently, nonlinear simulations are conducted for three case studies: conventional quadcopter, quadcopter-payload system without imperfections, quadcopter-payload system with imperfections. Finally, the results are presented demonstrating the effectiveness of the proposed control strategy.

Quickly finding high-quality paths is of great significance for autonomous underwater vehicles (AUVs) in path planning problems. In this paper, we present a cylinder-based heuristic rapidly exploring random tree (Cyl-HRRT*) algorithm, which is the extension version of the path planner presented in our previous publication. Cyl-HRRT* increases the likelihood of sampling states that can improve the current solution by biasing the states into a cylindrical subset, thus providing better paths for AUVs. A direct greedy sampling method is proposed to explore the space more efficiently and accelerate convergence to the optimum. To reasonably balance the optimization accuracy and the number of iterations, a beacon-based adaptive optimization strategy is presented, which adaptively establishes a cylindrical subset for the next focused sampling according to the current path. Furthermore, the Cyl-HRRT* algorithm is shown to be probabilistically complete and asymptotically optimal. Finally, the Cyl-HRRT* algorithm is comprehensively tested in both simulations and real-world experiments. The results reveal that the path generated by the Cyl-HRRT* algorithm greatly improves the power savings and mobility of the AUV.

Many people are fascinated by biological swarms, but understanding the behavior and inherent task objectives of a bird flock or ant colony requires training. Whereas several swarm intelligence works focus on mimicking natural swarm behaviors, we argue that this may not be the most intuitive approach to facilitate communication with the operators. Instead, we focus on the legibility of swarm expressive motions to communicate mission-specific messages to the operator. To do so, we leverage swarm intelligence algorithms on chain formation for resilient exploration and mapping combined with acyclic graph formation (AGF) into a novel swarm-oriented programming strategy. We then explore how expressive motions of robot swarms could be designed and test the legibility of nine different expressive motions in an online user study with 98 participants. We found several differences between the motions in communicating messages to the users. These findings represent a promising starting point for the design of legible expressive motions for implementation in decentralized robot swarms.

Designing and deploying autonomous swarms capable of performing collective tasks in real-world is extremely challenging. One drawback of getting out of the lab is that realistic tasks involve long distances with limited numbers of robots, leading to sparse and intermittent connectivity. As an example, search and rescue requires robots to coordinate in their search, and relay the information of found targets. The search’s effectiveness is greatly reduced if robots must stay close to maintain connectivity. This paper proposes a decentralized search system that only requires sporadic connectivity and allows information diffusion through the swarm whenever possible. Our robots share and update a distributed belief map, to coordinate the search. Once a target is detected, the robots form a communication relay between a base station and the target’s position. We show the applicability of our system both in simulation and with real-world experiments with a small swarm of drones.

Self-assembly of modular robotic systems enables the construction of complex robotic configurations to adapt to different tasks. This paper presents a framework for SMORES types of modular robots to efficiently self-assemble into tree topologies. These modular robots form kinematic chains that have been shown to be capable of a large variety of manipulation and locomotion tasks, yet they can reconfigure using a mobile reconfiguration. A desired kinematic topology can be mapped onto a planar pattern with the optimal module assignment based on the modules’ locations, then the mobile reconfiguration assembly process can be executed in parallel. A docking controller is developed to guarantee the success of docking processes. A hybrid control architecture is designed to handle a large number of modules and complex behaviors of each individual, and achieve efficient and robust self-assembly actions. The framework is demonstrated in both hardware and simulation on the SMORES-EP platform.