基于改进PSO的自主车辆实时运动规划方法

通过在孤长-曲率空间建立车辆运动学模型的方法,在满足非完整约束条件的基础上,将运动规划问题转化为函数优化问题.为提高PSO算法的优化速度,满足算法工程应用的实时性要求,提出一种基于多任务种群协同进化的粒子群优化算法.该算法将种群分为3种执行不同任务动作的子群,充分扩展搜索范围,挖掘搜索域内的有用信息,使种群的全局搜索能力和局部搜索能力达到较好的平衡状态.实验结果证明,将协同进化PSO算法应用于弧长-曲率空间中的函数优化问题,可实现对自主车辆的运动规划,规划轨迹满足车辆运动学和动力学约束,保证了车辆行驶的安全性和平稳性.     

基于Frenet坐标系的自动驾驶轨迹规划与优化算法

针对公路场景下作速度保持的自动驾驶汽车实时轨迹规划问题,提出一种基于Frenet坐标系的优化轨迹规划算法.首先,利用Frenet坐标系将车辆运动解耦,构建无约束横向/纵向独立积分系统;然后,根据初始配置和可内嵌到行为层的目标配置,通过采样生成有限的4次、5次多项式候选轨迹集合;最后,利用以高斯卷积、加速度变化率为核心的安全性和舒适性评价指标构建损失函数,评价轨迹成本,并结合曲率、加速度检查,选择能够最小化损失的最优解.结果表明,该算法能满足公路型场景的规划需求,车辆运动轨迹平滑、舒适、安全性更高.     

自动泊车运动规划方法综述

自动泊车是自动驾驶车辆在低速行驶场景中的核心技术之一,已被广泛应用于各类车型中。运动规划确保所规划出的轨迹能指引车辆从起始位姿无碰撞、符合运动学规律地行驶至目标位姿,是体现自动泊车智能水准的关键技术模块。随着当前自动驾驶技术从初步的技术原型研究逐渐向产业化应用转型,低速自动驾驶领域中的自动泊车技术成为了这一转型过程中的关键突破点。文章回顾了现有的应用于自动泊车领域的运动规划方法,将其归纳为几何方法、采样方法、搜索方法、数值优化方法以及机器学习方法等类别,并对各类方法的发展历程与技术特点进行了回顾;探讨了自动泊车运动规划方法的未来研究方向,具体包括构建泛化性更强的模型、提升规划方法的稳定性、关注在线实时高质量重规划的能力与防御性驾驶的能力,以及考虑与其他交通参与者的交互等。     

城市交叉口车辆速度与交通信号协同优化控制

为了降低城市交通中的行车延误与燃油消耗,针对人类驾驶车辆与自动驾驶车辆混合交通环境,提出一种基于交通信息物理系统(TCPS)的车辆速度与交通信号协同优化控制方法.首先,综合考虑路口交通信号、人类驾驶车辆、自动驾驶车辆三者之间的相互影响,设计一种适用于自动驾驶车辆与人类驾驶车辆混合组队特性的过路口速度规划模型;其次,针对车辆速度规划单一应用时的局限性,即无法减少车辆路口通行延误且易出现无解情况,提出一种双目标协同优化模型,能够综合考虑车辆速度规划与路口交通信号控制,同时降低车辆燃油消耗与路口平均延误.由于双目标优化问题求解的复杂性,设计一种遗传算法-粒子群算法混合求解策略.基于SUMO的仿真实验验证了所提出方法的有效性.     

基于贝塞尔曲线的智能汽车避障轨迹规划方法研究

开展了一种基于贝塞尔曲线的智能汽车避障局部轨迹规划,即路径规划和速度规划方法研究。路径规划时,为了适应各种形状道路,将道路笛卡尔坐标转换为Frénet坐标,以路径的长度、曲率和连续性,以及车辆碰撞风险为代价函数,其中引入危险势场理论,描述车辆碰撞风险,并采用序列二次规划方法来求解路径规划这一非线性优化问题;速度规划时,以行车效率和舒适性为目标,实现速度规划,该方法可以通过调整各子目标函数的权重来满足不同驾驶需求。为了验证基于贝塞尔曲线轨迹规划算法的有效性,设计了直道和弯道上静态和动态避障场景的仿真实验,结果表明,提出的轨迹规划方法能够在各种形状道路上完成避障任务,且避障过程中车辆状态变化平稳,能够保证乘坐舒适性。     

基于并行多种群自适应蚁群算法的聚类分析

数据聚类是数据挖掘中的一个重要课题。聚类问题可以归结为一个优化问题。蚁群算法作为一种鲁棒性很强的优化算法具有很强的全局优化能力。该文给出了一种并行多种群自适应蚁群算法。该算法采用多种群并行搜索,并在种群中采用基于目标函数值的启发式信息素分配策略和根据目标函数自动调整蚂蚁搜索路径的行为。理论分析和仿真实验表明,该算法是非常有效的。     

基于交互车辆轨迹预测的自动驾驶车辆轨迹规划

针对城市道路等复杂行车场景,提出了一种基于交互车辆轨迹预测的自动驾驶车辆轨迹规划方法,将高维度的轨迹规划解耦为低维度的路径规划和速度规划;首先,采用五次多项式曲线和碰撞剩余时间规划车辆行驶路径;其次,在社会生成对抗网络Social-GAN的基础上结合车辆空间影响和注意力机制对交互车辆进行轨迹预测;然后,结合主车规划路径、交互车辆预测轨迹及碰撞判定模型得到主车S-T图,采用动态规划和数值优化方法求解S-T图,从而得到满足车辆动力学约束的安全、舒适最优速度曲线;最后,搭建PreScan-CarSim-Matlab&Simulink-Python联合仿真模型进行实验验证。仿真结果表明,提出的轨迹规划方法能够在对交互车辆有效避撞的前提下,保证车辆行驶的舒适性和高效性。     

动态环境下改进蚁群算法的多Agent路径规划

针对动态环境下的多Agent路径规划问题,提出了一种改进的蚁群算法与烟花算法相结合的动态路径规划方法。通过自适应信息素强度值及信息素缩减因子来加快算法的迭代速度,并利用烟花算法来解决路径规划过程中的死锁问题,避免陷入局部最优。在多Agent动态避碰过程中,根据动态障碍物与多Agent之间的运行轨迹是否相交制定相应的避碰策略,并利用路径转变函数解决多Agent的正面碰撞问题。仿真实验表明,该方法优于经典蚁群算法,能够有效解决多Agent路径规划中的碰撞问题,从而快速找到最优无碰路径。     

基于k度平滑的多无人机协调路径规划方法

针对多无人机协同的轨迹规划问题,提出一种基于k度平滑法的多无人机协同路径规划方法.通过改进的蚁群优化算法搜索最短路径,应用k度平滑方法平滑初始路径并实现多无人机的协调.通过k度平滑引入多无人机协调算法,使多架无人机能够在k度时间间隔内到达指定地点.通过仿真实验将所提方法与改进蚁群算法、经典算法及平滑方法进行对比分析,验证所提算法的可行性和有效性.实验结果表明,所提策略能够有效提高多无人机路径规划效率,实现多无人机同时到达指定位置的目标,且具有良好的可行性、有效性.     

煤矿井下移动机器人运动规划方法研究

针对现有煤矿井下移动机器人运动规划所生成的轨迹存在超调、碰撞、不连续、不光滑等问题,提出了一种由路径规划、轨迹生成、轨迹优化3个部分构成的煤矿井下移动机器人运动规划方法。路径规划采用基于图搜索的A*算法实现,通过开始搜索、路径排序、继续搜索3个步骤循环迭代,快速规划出一条可通行的全局路径作为轨迹生成的初值。轨迹生成通过构建基于Minimum Snap的目标函数,并施加等式约束来实现。轨迹优化则是在轨迹生成的基础上施加不等式约束来实现:通过调整时间分配和构建基于Corridor轨迹规划的不等式约束,解决基于Minimum Snap轨迹生成在求解过程中出现的超调现象,并对整段轨迹本身进行约束,避免发生碰撞;通过引入调和函数Bezier Curve,构建基于Bezier Curve的Minimum Snap的轨迹优化问题,使得轨迹高阶目标函数的求解变得简单高效,最终生成一条适用于煤矿井下移动机器人的能量损失最小、连续、光滑、无碰撞、可执行的运动轨迹。在Matlab仿真环境中设计了随机地图,生成了包含时间分配、位置规划、速度规划、加速度规划的最优轨迹规划结果。实验结果验证了该运动规划方法的正确性和有效性。     

Receding-Horizon Trajectory Planning for Under-Actuated Autonomous Vehicles Based on Collaborative Neurodynamic Optimization

This paper addresses a major issue in planning the trajectories of under-actuated autonomous vehicles based on neurodynamic optimization. A receding-horizon vehicle trajectory planning task is formulated as a sequential global optimization problem with weighted quadratic navigation functions and obstacle avoidance constraints based on given vehicle goal configurations. The feasibility of the formulated optimization problem is guaranteed under derived conditions. The optimization problem is sequentially solved via collaborative neurodynamic optimization in a neurodynamics-driven trajectory planning method/procedure. Simulation results with under-actuated unmanned wheeled vehicles and autonomous surface vehicles are elaborated to substantiate the efficacy of the neurodynamics-driven trajectory planning method.      

An online collision-free trajectory generation algorithm for human–robot collaboration

The premise of human–robot collaboration is that robots have adaptive trajectory planning strategies in hybrid work cell. The aim of this paper is to propose a new online collision avoidance trajectory planning algorithm for moderate dynamic environments to insure human safety when sharing collaborative tasks. The algorithm contains two parts: trajectory generation and local optimization. Firstly, based on empirical Dirichlet Process Gaussian Mixture Model (DPGMM) distribution learning, a neural network trajectory planner called Collaborative Waypoint Planning network (CWP-net) is proposed to generate all key waypoints required for dynamic obstacle avoidance in joint space according to environmental inputs. These points are used to generate quintic spline smooth motion trajectories with velocity and acceleration constraints. Secondly, we present an improved Stochastic Trajectory Optimization for Motion Planning (STOMP) algorithm which locally optimizes the generated trajectories of CWP-net by constraining the trajectory optimization range and direction through the DPGMM model. Simulations and real experiments from an industrial use case of human–robot collaboration in the field of aircraft assembly testing show that the proposed algorithm can smoothly adjust the nominal path online and effectively avoid collisions during the collaboration.     

Autonomous driving at the handling limit using residual reinforcement learning

While driving a vehicle safely at its handling limit is essential in autonomous vehicles in Level 5 autonomy, it is a very challenging task for current conventional methods. Therefore, this study proposes a novel controller of trajectory planning and motion control for autonomous driving through manifold corners at the handling limit to improve the speed and shorten the lap time of the vehicle. The proposed controller innovatively combines the advantages of conventional model-based control algorithm, model-free reinforcement learning algorithm, and prior expert knowledge, to improve the training efficiency for autonomous driving in extreme conditions. The reward shaping of this algorithm refers to the procedure and experience of race training of professional drivers in real time. After training on track maps that exhibit different levels of difficulty, the proposed controller implemented a superior strategy compared to the original reference trajectory, and can to other tougher maps based on the basic driving knowledge learned from the simpler map, which verifies its superiority and extensibility. We believe this technology can be further applied to daily life to expand the application scenarios and maneuvering envelopes of autonomous vehicles.     

General fight rule-based trajectory planning for pairwise collision avoidance in a known environment

This paper presents a general flight rule-based autonomous trajectory planning scheme for two aerial vehicles to avoid obstacles and collisions in known environments in low-altitude airspace for general aviation. Flight rules in low-altitude airspace are first introduced based on the general flight rules in US, UK and China, and then the suitable flight rules are embedded into the trajectory planning algorithm. It is supposed that the flight parameters, such as positions and velocities, are all available to the aerial vehicles involved in the possible conflict. Then the trajectory of each aerial vehicle is calculated by optimizing an objective function, such as distance and fuel consumption, with the constraints corresponding to the airspace traffic rules. The optimization problem is solved by receding horizon control (RHC) based mixed integer linear programming (MILP). Compared with other collision avoidance algorithms, the proposed algorithm can be adapted to plan the autonomous trajectory to avoid pairwise collision and obstacles as proposed general flight rules. Simulations show the feasibility of the proposed scheme.     

基于改进人工鱼群算法的车辆轨迹规划方法

针对车联网环境下若干典型车辆轨迹规划方法存在车速与轨迹波动性较大的问题,提出一种基于改进人工鱼群算法的车辆轨迹规划方法。该方法以短程通信（DSRC）的车联网应用场景为设计平台,以车辆的最优行车速度为核心计算基础,分析得到了车辆的最佳轨迹。首先,对人工鱼群算法在车联网应用场景的优势和不足进行分析,引入万有引力力学模型与避障模式控制,提出一种改进的人工鱼群算法;然后,分析车辆在车联网应用场景中的受力约束,利用网联车辆的自组织行为控制策略推导最优行车速度;最后,基于最优行车速度实现对车辆的实时轨迹诱导和轨迹避障控制规划。仿真测试结果表明,在运用了基于改进人工鱼群算法的轨迹规划模型后,车辆的驾驶速度更加平稳,轨迹波动性较小,对障碍物可实现零失误避撞;在多车相遇情况下,测试车辆为2~40时,相对于原人工鱼群算法和萤火虫算法,运用改进人工鱼群算法后车速的平均迭代次数减少,迭代效率提高3~7、4~8倍,且随着车辆数目越多,迭代效率提升越明显。     

Autonomous Navigation of UAV in Foliage Environment

This paper presents a navigation system that enables small-scale unmanned aerial vehicles to navigate autonomously using a 2D laser range finder in foliage environment without GPS. The navigation framework consists of real-time dual layer control, navigation state estimation and online path planning. In particular, the inner loop of a quadrotor is stabilized using a commercial autopilot while the outer loop control is implemented using robust perfect tracking. The navigation state estimation consists of real-time onboard motion estimation and trajectory smoothing using the GraphSLAM technique. The onboard real-time motion estimation is achieved by a Kalman filter, fusing the planar velocity measurement from matching the consecutive scans of a laser range finder and the acceleration measurement of an inertial measurement unit. The trajectory histories from the real-time autonomous navigation together with the observed features are fed into a sliding-window based pose-graph optimization framework. The online path planning module finds an obstacle-free trajectory based the local measurement of the laser range finder. The performance of the proposed navigation system is demonstrated successfully on the autonomous navigation of a small-scale UAV in foliage environment.     

On the Generation of Trajectories for Multiple UAVs in Environments with Obstacles

This paper presents a methodology based on a variation of the Rapidly-exploring Random Trees (RRTs) that generates feasible trajectories for a team of autonomous aerial vehicles with holonomic constraints in environments with obstacles. Our approach uses Pythagorean Hodograph (PH) curves to connect vertices of the tree, which makes it possible to generate paths for which the main kinematic constraints of the vehicle are not violated. These paths are converted into trajectories based on feasible speed profiles of the robot. The smoothness of the acceleration profile of the vehicle is indirectly guaranteed between two vertices of the RRT tree. The proposed algorithm provides fast convergence to the final trajectory. We still utilize the properties of the RRT to avoid collisions with static, environment bound obstacles and dynamic obstacles, such as other vehicles in the multi-vehicle planning scenario. We show results for a set of small unmanned aerial vehicles in environments with different configurations.     

求解多车辆装载问题的启发式改进蚁群算法设计

针对启发式优化算法不能较理想地对多车辆大规模装载问题进行优化的局限性,文章设计了一种启发式改进蚁群算法,该算法将单车辆的启发式装载与多车辆装载时的蚁群优化算法有机结合,较好地解决了多车辆大规模装载问题。经过实例验证,该算法具有较高的计算效率和较好的收敛特性。     

Time-optimal and Smooth Trajectory Planning for Robot Manipulators

This paper presents a practical time-optimal and smooth trajectory planning algorithm and then applies it to robot manipulators. The proposed algorithm uses the time-optimal theory based on the dynamics model to plan the robot’s motion trajectory, constructs the trajectory optimization model under the constraints of the geometric path and joint torque, and dynamically selects the optimal trajectory parameters during the solving process to prominently improve the robot’s motion speed. Moreover, the proposed algorithm utilizes the input shaping algorithm instead of the jerk constraint in the trajectory optimization model to achieve a smooth trajectory. The input shaping of trajectory parameters during postprocessing not only suppresses the residual vibration of the robot but also takes the signal delay caused by traditional input shaping into account. The combination of these algorithms makes the proposed time-optimal and smooth trajectory planning algorithm ensure absolute time optimality and achieve a smooth trajectory. The results of an experiment on a six-degree-of-freedom industrial robot indicate the validity of the proposed algorithm.