Due to the requirements for mobile robots to search or rescue in unknown environments,reactive navigation which plays an essential role in these applications has attracted increasing interest.However,most existing rea...Due to the requirements for mobile robots to search or rescue in unknown environments,reactive navigation which plays an essential role in these applications has attracted increasing interest.However,most existing reactive methods are vulnerable to local minima in the absence of prior knowledge about the environment.This paper aims to address the local minimum problem by employing the proposed boundary gap(BG)based reactive navigation method.Specifically,the narrowest gap extraction algorithm(NGEA)is proposed to eliminate the improper gaps.Meanwhile,we present a new concept called boundary gap which enables the robot to follow the obstacle boundary and then get rid of local minima.Moreover,in order to enhance the smoothness of generated trajectories,we take the robot dynamics into consideration by using the modified dynamic window approach(DWA).Simulation and experimental results show the superiority of our method in avoiding local minima and improving the smoothness.展开更多
This paper presents a new algorithm of path planning for mobile robots,which utilises the characteristics of the obstacle border and fuzzy logical reasoning.The environment topology or working space is described by th...This paper presents a new algorithm of path planning for mobile robots,which utilises the characteristics of the obstacle border and fuzzy logical reasoning.The environment topology or working space is described by the time-variable grid method that can be further described by the moving obstacles and the variation of path safety.Based on the algorithm,a new path planning approach for mobile robots in an unknown environment has been developed.The path planning approach can let a mobile robot find a safe path from the current position to the goal based on a sensor system.The two types of machine learning:advancing learning and exploitation learning or trial learning are explored,and both are applied to the learning of mobile robot path planning algorithm.Comparison with A*path planning approach and various simulation results are given to demonstrate the efficiency of the algorithm.This path planning approach can also be applied to computer games.展开更多
Unmanned aerial vehicles(UAVs) are increasingly considered in safe autonomous navigation systems to explore unknown environments where UAVs are equipped with multiple sensors to perceive the surroundings. However, how...Unmanned aerial vehicles(UAVs) are increasingly considered in safe autonomous navigation systems to explore unknown environments where UAVs are equipped with multiple sensors to perceive the surroundings. However, how to achieve UAVenabled data dissemination and also ensure safe navigation synchronously is a new challenge. In this paper, our goal is minimizing the whole weighted sum of the UAV’s task completion time while satisfying the data transmission task requirement and the UAV’s feasible flight region constraints. However, it is unable to be solved via standard optimization methods mainly on account of lacking a tractable and accurate system model in practice. To overcome this tough issue,we propose a new solution approach by utilizing the most advanced dueling double deep Q network(dueling DDQN) with multi-step learning. Specifically, to improve the algorithm, the extra labels are added to the primitive states. Simulation results indicate the validity and performance superiority of the proposed algorithm under different data thresholds compared with two other benchmarks.展开更多
Compliant interaction control is a key technology for robots performing contact-rich manipulation tasks.The design of the compliant controller needs to consider the robot hardware because complex control algorithms ma...Compliant interaction control is a key technology for robots performing contact-rich manipulation tasks.The design of the compliant controller needs to consider the robot hardware because complex control algorithms may not be compatible with the hardware performance,especially for some industrial robots with low bandwidth sensors.This paper focuses on effective and easy-to-use compliant control algorithms for position/velocity-controlled robots.Inspired by human arm stiffness adaptation behavior,a novel variable target stiffness(NVTS)admittance control strategy is proposed for adaptive force tracking,in which a proportional integral derivative(PID)variable stiffness law is designed to update the stiffness coefficient of the admittance function by the force and position feedback.Meanwhile,its stability and force-tracking capability are theoretically proven.In addition,an impact compensator(Impc)is integrated into the NVTS controller to enhance its disturbance-suppression capability when the robot is subjected to strong vibration disturbances in complicated surface polishing tasks.The proposed controllers are validated through four groups of experimental tests using different robots and the corresponding results demonstrate that they have high-accuracy tracking capability and strong adaptability in unknown environments.展开更多
This paper investigates the challenges associated with Unmanned Aerial Vehicle (UAV) collaborative search and target tracking in dynamic and unknown environments characterized by limited field of view. The primary obj...This paper investigates the challenges associated with Unmanned Aerial Vehicle (UAV) collaborative search and target tracking in dynamic and unknown environments characterized by limited field of view. The primary objective is to explore the unknown environments to locate and track targets effectively. To address this problem, we propose a novel Multi-Agent Reinforcement Learning (MARL) method based on Graph Neural Network (GNN). Firstly, a method is introduced for encoding continuous-space multi-UAV problem data into spatial graphs which establish essential relationships among agents, obstacles, and targets. Secondly, a Graph AttenTion network (GAT) model is presented, which focuses exclusively on adjacent nodes, learns attention weights adaptively and allows agents to better process information in dynamic environments. Reward functions are specifically designed to tackle exploration challenges in environments with sparse rewards. By introducing a framework that integrates centralized training and distributed execution, the advancement of models is facilitated. Simulation results show that the proposed method outperforms the existing MARL method in search rate and tracking performance with less collisions. The experiments show that the proposed method can be extended to applications with a larger number of agents, which provides a potential solution to the challenging problem of multi-UAV autonomous tracking in dynamic unknown environments.展开更多
Flapping Wing Aerial Vehicles(FWAVs)hold immense potential for applications such as search-and-rescue missions in complex terrains,environmental monitoring in hazardous areas,and exploration in confined spaces.However...Flapping Wing Aerial Vehicles(FWAVs)hold immense potential for applications such as search-and-rescue missions in complex terrains,environmental monitoring in hazardous areas,and exploration in confined spaces.However,their adoption is hindered by the challenges of autonomous navigation in unknown environments,exacerbated by their limited onboard computational resources and demanding flight dynamics.This work addresses these challenges by presenting a lightweight,vision-based autonomous navigation system weighing 26.0 g,enabling FWAVs to achieve obstacle-avoidance flight at a speed of 9.0 m/s.Central to this system is a novel end-toend Bi-level Cooperative Policy(BCP)that significantly improves flight efficiency and safety.BCP employs lightweight neural networks for real-time performance and leverages Hierarchical Reinforcement Learning(HRL)for robust and efficient training.Quantitative evaluations show that BCP achieves up to 6.5%shorter path lengths,11.2%faster task completion time,and improved explainability compared to state-of-the-art reinforcement learning algorithms.Additionally,BCP demonstrates 35.7%more efficient and stable training,reducing computational overhead while maintaining high performance.The system design incorporates optimized lightweight components,including a 4.0 g customized stereo camera,a 6.0 g 3D-printed camera mount,and a 16.0 g onboard computer,all tailored to FWAV applications.Real-flight experiments validate the sim-toreal transferability of the proposed navigation system,demonstrating its readiness for real-world deployment in challenging scenarios.This research advances the practicality of FWAVs,paving the way for their broader adoption in critical missions where compact,agile aerial robots are indispensable.展开更多
This study investigates robot path planning for multiple agents,focusing on the critical requirement that agents can pursue concurrent pathways without collisions.Each agent is assigned a task within the environment t...This study investigates robot path planning for multiple agents,focusing on the critical requirement that agents can pursue concurrent pathways without collisions.Each agent is assigned a task within the environment to reach a designated destination.When the map or goal changes unexpectedly,particularly in dynamic and unknown environments,it can lead to potential failures or performance degradation in various ways.Additionally,priority inheritance plays a significant role in path planning and can impact performance.This study proposes a ConflictBased Search(CBS)approach,introducing a unique hierarchical search mechanism for planning paths for multiple robots.The study aims to enhance flexibility in adapting to different environments.Three scenarios were tested,and the accuracy of the proposed algorithm was validated.In the first scenario,path planning was applied in unknown environments,both stationary and mobile,yielding excellent results in terms of time to arrival and path length,with a time of 2.3 s.In the second scenario,the algorithm was applied to complex environments containing sharp corners and unknown obstacles,resulting in a time of 2.6 s,with the algorithm also performing well in terms of path length.In the final scenario,the multi-objective algorithm was tested in a warehouse environment containing fixed,mobile,and multi-targeted obstacles,achieving a result of up to 100.4 s.Based on the results and comparisons with previous work,the proposed method was found to be highly effective,efficient,and suitable for various environments.展开更多
针对环境遮挡与交通参与者行为随机导致的驾驶风险,提出一种面向无信号灯十字路口场景的安全决策方法。首先,建立一种基于值分布式强化学习-全参数化分位数网络(Fully parameterized quantile network,FPQN)的基础决策策略。其次,融合F...针对环境遮挡与交通参与者行为随机导致的驾驶风险,提出一种面向无信号灯十字路口场景的安全决策方法。首先,建立一种基于值分布式强化学习-全参数化分位数网络(Fully parameterized quantile network,FPQN)的基础决策策略。其次,融合FPQN建模的累积回报分布与条件风险价值函数(Conditional value at risk,CVaR),进而构建具有驾驶风险意识的安全决策策略。再次,引入集成学习理论(Ensemble),建立基于集成FPQN的决策不确定性估计框架EFPQN,能够实时量化决策风险。同时,为应对决策不确定性较高带来的驾驶风险,设计基于模型预测控制的备选策略以提升安全性。最后,采用SUMO仿真平台搭建无信号灯十字路口场景,对提出的安全决策方法进行验证。试验结果表明,与基准方法相比,所提出的方法能够有效降低遮挡与交通参与者行为随机导致的驾驶风险。展开更多
This paper proposes a multi-UAV cooperative exploration approach based on task-density space partition.In the research of multi-UAV cooperative exploration,it is a prevalent cooperative scheme to control robots to wor...This paper proposes a multi-UAV cooperative exploration approach based on task-density space partition.In the research of multi-UAV cooperative exploration,it is a prevalent cooperative scheme to control robots to work independently in partitioned spaces.Nonetheless,only considering the position of robots during space partition cannot effectively ensure the overall cooperative efficiency.According to research on task density of current time points and positions of robots during exploration,robots with fewer task points are assigned to work in spaces with more tasks in the rolling horizon optimization planning mode,which can reduce the redundancy of multi-robot cooperative work.Comparative research suggests that the overall exploration efficiency is improved.展开更多
基金supported in part by the National Natural Science Foundation of China(61922076,61873252)in part by the Fok Ying-Tong Education Foundation for Young Teachers in Higher Education Institutions of China(161059)。
文摘Due to the requirements for mobile robots to search or rescue in unknown environments,reactive navigation which plays an essential role in these applications has attracted increasing interest.However,most existing reactive methods are vulnerable to local minima in the absence of prior knowledge about the environment.This paper aims to address the local minimum problem by employing the proposed boundary gap(BG)based reactive navigation method.Specifically,the narrowest gap extraction algorithm(NGEA)is proposed to eliminate the improper gaps.Meanwhile,we present a new concept called boundary gap which enables the robot to follow the obstacle boundary and then get rid of local minima.Moreover,in order to enhance the smoothness of generated trajectories,we take the robot dynamics into consideration by using the modified dynamic window approach(DWA).Simulation and experimental results show the superiority of our method in avoiding local minima and improving the smoothness.
文摘This paper presents a new algorithm of path planning for mobile robots,which utilises the characteristics of the obstacle border and fuzzy logical reasoning.The environment topology or working space is described by the time-variable grid method that can be further described by the moving obstacles and the variation of path safety.Based on the algorithm,a new path planning approach for mobile robots in an unknown environment has been developed.The path planning approach can let a mobile robot find a safe path from the current position to the goal based on a sensor system.The two types of machine learning:advancing learning and exploitation learning or trial learning are explored,and both are applied to the learning of mobile robot path planning algorithm.Comparison with A*path planning approach and various simulation results are given to demonstrate the efficiency of the algorithm.This path planning approach can also be applied to computer games.
基金supported by the National Natural Science Foundation of China (No. 61931011)。
文摘Unmanned aerial vehicles(UAVs) are increasingly considered in safe autonomous navigation systems to explore unknown environments where UAVs are equipped with multiple sensors to perceive the surroundings. However, how to achieve UAVenabled data dissemination and also ensure safe navigation synchronously is a new challenge. In this paper, our goal is minimizing the whole weighted sum of the UAV’s task completion time while satisfying the data transmission task requirement and the UAV’s feasible flight region constraints. However, it is unable to be solved via standard optimization methods mainly on account of lacking a tractable and accurate system model in practice. To overcome this tough issue,we propose a new solution approach by utilizing the most advanced dueling double deep Q network(dueling DDQN) with multi-step learning. Specifically, to improve the algorithm, the extra labels are added to the primitive states. Simulation results indicate the validity and performance superiority of the proposed algorithm under different data thresholds compared with two other benchmarks.
基金the National Natural Science Foundation of China(Grant Nos.62103407,52075530,and 52175272)the State Key Laboratory of Robotics Foundation(Grant No.Y91Z0303)。
文摘Compliant interaction control is a key technology for robots performing contact-rich manipulation tasks.The design of the compliant controller needs to consider the robot hardware because complex control algorithms may not be compatible with the hardware performance,especially for some industrial robots with low bandwidth sensors.This paper focuses on effective and easy-to-use compliant control algorithms for position/velocity-controlled robots.Inspired by human arm stiffness adaptation behavior,a novel variable target stiffness(NVTS)admittance control strategy is proposed for adaptive force tracking,in which a proportional integral derivative(PID)variable stiffness law is designed to update the stiffness coefficient of the admittance function by the force and position feedback.Meanwhile,its stability and force-tracking capability are theoretically proven.In addition,an impact compensator(Impc)is integrated into the NVTS controller to enhance its disturbance-suppression capability when the robot is subjected to strong vibration disturbances in complicated surface polishing tasks.The proposed controllers are validated through four groups of experimental tests using different robots and the corresponding results demonstrate that they have high-accuracy tracking capability and strong adaptability in unknown environments.
基金supported by the National Natural Science Foundation of China(Nos.12272104,U22B2013).
文摘This paper investigates the challenges associated with Unmanned Aerial Vehicle (UAV) collaborative search and target tracking in dynamic and unknown environments characterized by limited field of view. The primary objective is to explore the unknown environments to locate and track targets effectively. To address this problem, we propose a novel Multi-Agent Reinforcement Learning (MARL) method based on Graph Neural Network (GNN). Firstly, a method is introduced for encoding continuous-space multi-UAV problem data into spatial graphs which establish essential relationships among agents, obstacles, and targets. Secondly, a Graph AttenTion network (GAT) model is presented, which focuses exclusively on adjacent nodes, learns attention weights adaptively and allows agents to better process information in dynamic environments. Reward functions are specifically designed to tackle exploration challenges in environments with sparse rewards. By introducing a framework that integrates centralized training and distributed execution, the advancement of models is facilitated. Simulation results show that the proposed method outperforms the existing MARL method in search rate and tracking performance with less collisions. The experiments show that the proposed method can be extended to applications with a larger number of agents, which provides a potential solution to the challenging problem of multi-UAV autonomous tracking in dynamic unknown environments.
基金supported by the Fundamental Research Funds for the Central Universities,China。
文摘Flapping Wing Aerial Vehicles(FWAVs)hold immense potential for applications such as search-and-rescue missions in complex terrains,environmental monitoring in hazardous areas,and exploration in confined spaces.However,their adoption is hindered by the challenges of autonomous navigation in unknown environments,exacerbated by their limited onboard computational resources and demanding flight dynamics.This work addresses these challenges by presenting a lightweight,vision-based autonomous navigation system weighing 26.0 g,enabling FWAVs to achieve obstacle-avoidance flight at a speed of 9.0 m/s.Central to this system is a novel end-toend Bi-level Cooperative Policy(BCP)that significantly improves flight efficiency and safety.BCP employs lightweight neural networks for real-time performance and leverages Hierarchical Reinforcement Learning(HRL)for robust and efficient training.Quantitative evaluations show that BCP achieves up to 6.5%shorter path lengths,11.2%faster task completion time,and improved explainability compared to state-of-the-art reinforcement learning algorithms.Additionally,BCP demonstrates 35.7%more efficient and stable training,reducing computational overhead while maintaining high performance.The system design incorporates optimized lightweight components,including a 4.0 g customized stereo camera,a 6.0 g 3D-printed camera mount,and a 16.0 g onboard computer,all tailored to FWAV applications.Real-flight experiments validate the sim-toreal transferability of the proposed navigation system,demonstrating its readiness for real-world deployment in challenging scenarios.This research advances the practicality of FWAVs,paving the way for their broader adoption in critical missions where compact,agile aerial robots are indispensable.
文摘This study investigates robot path planning for multiple agents,focusing on the critical requirement that agents can pursue concurrent pathways without collisions.Each agent is assigned a task within the environment to reach a designated destination.When the map or goal changes unexpectedly,particularly in dynamic and unknown environments,it can lead to potential failures or performance degradation in various ways.Additionally,priority inheritance plays a significant role in path planning and can impact performance.This study proposes a ConflictBased Search(CBS)approach,introducing a unique hierarchical search mechanism for planning paths for multiple robots.The study aims to enhance flexibility in adapting to different environments.Three scenarios were tested,and the accuracy of the proposed algorithm was validated.In the first scenario,path planning was applied in unknown environments,both stationary and mobile,yielding excellent results in terms of time to arrival and path length,with a time of 2.3 s.In the second scenario,the algorithm was applied to complex environments containing sharp corners and unknown obstacles,resulting in a time of 2.6 s,with the algorithm also performing well in terms of path length.In the final scenario,the multi-objective algorithm was tested in a warehouse environment containing fixed,mobile,and multi-targeted obstacles,achieving a result of up to 100.4 s.Based on the results and comparisons with previous work,the proposed method was found to be highly effective,efficient,and suitable for various environments.
文摘针对环境遮挡与交通参与者行为随机导致的驾驶风险,提出一种面向无信号灯十字路口场景的安全决策方法。首先,建立一种基于值分布式强化学习-全参数化分位数网络(Fully parameterized quantile network,FPQN)的基础决策策略。其次,融合FPQN建模的累积回报分布与条件风险价值函数(Conditional value at risk,CVaR),进而构建具有驾驶风险意识的安全决策策略。再次,引入集成学习理论(Ensemble),建立基于集成FPQN的决策不确定性估计框架EFPQN,能够实时量化决策风险。同时,为应对决策不确定性较高带来的驾驶风险,设计基于模型预测控制的备选策略以提升安全性。最后,采用SUMO仿真平台搭建无信号灯十字路口场景,对提出的安全决策方法进行验证。试验结果表明,与基准方法相比,所提出的方法能够有效降低遮挡与交通参与者行为随机导致的驾驶风险。
文摘This paper proposes a multi-UAV cooperative exploration approach based on task-density space partition.In the research of multi-UAV cooperative exploration,it is a prevalent cooperative scheme to control robots to work independently in partitioned spaces.Nonetheless,only considering the position of robots during space partition cannot effectively ensure the overall cooperative efficiency.According to research on task density of current time points and positions of robots during exploration,robots with fewer task points are assigned to work in spaces with more tasks in the rolling horizon optimization planning mode,which can reduce the redundancy of multi-robot cooperative work.Comparative research suggests that the overall exploration efficiency is improved.
