基于Sarsa(λ)强化学习的空间机械臂路径规划研究

针对目标特性未知的在轨操作环境,研究了典型空间操作机械臂的路径规划策略。采用Sarsa(λ)强化学习方法实现目标跟踪及避障的自主路径规划与智能决策,该方法将机械臂系统的每节臂视为一个决策智能体,通过感知由目标偏差和障碍距离程度组成的二维状态,设计符合人工经验的拟合奖赏函数,进行各臂转动动作的强化训练,最终形成各智能体的状态-动作值函数表,即可作为机械臂在线路径规划的决策依据。将本方法应用于多自由度空间机械臂路径规划任务,仿真结果表明新算法能在有限训练次数内实现对移动目标的稳定跟踪与避障,同时各智能体通过学习所得的状态-动作值函数表,具备较强的后期在线自主调整能力,从而验证了算法较强的鲁棒性和智能性。     

Variable Granularity in Route Directions

Abstract

This paper addresses changes of spatial granularity in route directions in relation to information needs in multimodal traveling. We outline a model of variability in granularity and apply this model to empirical data. Results reveal that linguistic route directions produced by humans as well as automatically generated web-based services provide the most crucial route elements in hierarchically structured ways that reflect the salient structure imposed by multimodal traveling. However, although the web-based information is impressively comprehensive, human route directions exhibit more flexibility regarding switches of place-related granularity, and they provide more detailed information at complex locations or decision points.     

Learning Routes from Visualizations for Indoor Wayfinding: Presentation Modes and Individual Differences

Abstract

Participants (N = 78) studied a visualization of a route through a complex building and walked that route in the real building without further assistance. Erroneous turns on the route as well as indicators of uncertainty such as hesitations were assessed. Three types of route visualizations were compared: (1) an allocentric, map-based visualization with the route indicated in floor maps, (2) an ordered sequence of pictures of decision points shown from the egocentric perspective, and (3) an animation showing a virtual walk of the route from the egocentric perspective. In addition to the experimental variation, gender differences, differences in visual-spatial abilities and differences in self-reported wayfinding strategies were considered as predictor variables. Wayfinding performance did not differ between allocentric (map) and egocentric (decision point pictures and animation) visualizations. However, wayfinding performance was better with animated than with static egocentric visualizations. Individual differences in the ability to encode visual-spatial information from the visualization played a critical role for route learning. Self-reported sense of direction related to egocentric wayfinding strategies also predicted wayfinding performance. Gender differences were attributable to differences in visual-spatial abilities and egocentric wayfinding strategies. Interactions between visualizations and individual differences were not found. It is concluded that animations of virtual walks are suitable to convey route information in complex buildings. Successful acquisition of route knowledge from maps is possible but might depend on the comprehensibility of the structure of the building.     

Dysfunctions of Spatial Cognition in Schizophrenic Patients

ABSTRACT

Twenty outpatients who fulfilled the criteria for a diagnosis of schizophrenia and 28 control participants were invited to learn a route through a complex outdoor environment. They were then tested in tasks intended to explore various aspects of their memorized representation of the navigational episode. Compared to controls, the patients showed significant impairment in both the verbal production of route directions and the drawing of sketch maps. They referred to fewer landmarks and provided fewer directional instructions than the controls, while making a greater number of irrelevant comments. When invited to distinguish between photographs showing views of landmarks encountered along the route and distractors, they performed as well as the controls, and they had similar response times. However, when they were presented with pairs of actual photographs taken along the route, they displayed special difficulty in deciding which of the two landmarks was encountered first along the route. This difficulty in retrieving the sequential structure of the navigational episode suggests that the patients' memories were not accurately linked to one another in their mental representation of the route. These findings are interpreted in the context of current hypotheses about the hippocampal impairment that affects schizophrenic patients.     

飞行器低空突防中的威胁航线优化技术研究

低空突防中的威胁回避技术是TF／TA^2系统的核心，直接关系到低空突防飞行器的任务生存能力。针对威胁回避的特殊要求，本文提出了一种基于地形可视性分析的威胁危险指标快速计算方法，解决了威胁的量化问题。针对在大范围任务区域内进行威胁航线优化存在计算复杂性和收敛性等问题，本文尝试采用遗传算法对威胁进行优化处理。该方法得到的威胁航线，严格经过出发点和目标点，且尽量远离威胁，有效提高了飞行器低空突防的任务生存率。     

动态环境中的飞行器实时三维航迹规划方法研究

本文提出了一种新的飞行器三维航迹规划方法－SDS。该方法能够在具有预先未知威胁的飞行环境中在线实时航迹规划。当飞行器上装备的探测器探测到飞行环境中有预先未知的威胁出现时，根据探测到的信息及时更新威胁数据。SDS根据新的环境信息局部修正受到影响的航迹段来获得新的全局最优航迹。在每一时刻考虑当前已知的信息下，SDS生成的航迹是满足要求的最优航迹。     

The Influence of Landmarks and Pre-exposure to a Structural Map During the Process of Spatial Knowledge Acquisition: A Study with Children and Adults in a Virtual Environment

ABSTRACT

This study investigated the effects of featural information (landmarks) and geometric information (pre-exposure to a structural map) and their possible interaction during the process of spatial knowledge acquisition of 8- and 11-year-old children and adults in a virtual environment. The study confirmed the well-known result of a developmental achievement in spatial cognition from childhood to adulthood. Although landmarks and the pre-exposure to a structural map did not affect the time to learn a specific route, they influenced the use of behavior in spatial learning and eased the acquisition of spatial knowledge measured by a route reversal and map-drawing tasks. Children and adults are able to integrate featural and geometric information in the spatial knowledge acquisition process in an environmental space, but their integration depends on the spatial processing stages that are investigated. Moreover, it was successfully demonstrated that the use of desktop virtual environments seems to be appropriate to investigate the development of spatial cognition.     

Spatial Influence of Environmental Axes in a Baseball Field

Abstract

We investigated the influence of environmental axes in a baseball field. In Experiment 1, participants walked either a path in the prototypical orientation (home plate to second base) or one which was rotated 225°. Recall for object locations was best when participants imagined themselves aligned with axes salient from the experienced orientation. In Experiment 2, when learning was through a route text, there was less of an influence of environmental axes. In Experiment 3, when participants walked both paths, memories were good for the atypical orientation, suggesting that task-specific spatial cues can be more influential than a prior conceptual north.     

Evidence for Amodal Representations after Bimodal Learning: Integration of Haptic-Visual Layouts into a Common Spatial Image

Abstract

Participants learned circular layouts of six objects presented haptically or visually, then indicated the direction from a start target to an end target of the same or different modality (intramodal versus intermodal). When objects from the two modalities were learned separately, superior performance for intramodal trials indicated a cost of switching between modalities. When a bimodal layout intermixing modalities was learned, intra- and intermodal trials did not differ reliably. These findings indicate that a spatial image, independent of input modality, can be formed when inputs are spatially and temporally congruent, but not when modalities are temporally segregated in learning.     

基于波碰撞法的机器人路径规划法

文章提出了在离散区域上,基于起始点和目标点相向传播波碰撞法的机器人路径规划法。设机器人作业环境为二维平面大小一定的矩形地形,机器人行走时仅限于作原地转向或直线行走。基于此开展路径规划的仿真研究,提出了搜索许可路径的算法。计算机模拟结果证实该算法可行、有效。     

一种三维航迹快速搜索方法

本文提出了基于SAS的自动三维航迹规划方法。该方法通过把约束条件结合到搜索算法中去，有效地减小了搜索空间，缩短了搜索时间，从而使三维规划能够用于实时航迹规划。在搜索过程中地形信息得到了充分利用，使算法生成的航迹能够自动回避地形和威胁。实验证明，该方法能够快速有效地完成规划任务，获得满意的三维航迹。     

Interface Familiarity Restores Active Advantage in a Virtual Exploration and Reconstruction Task in Children

Abstract

Active exploration is reportedly better than passive observation of spatial displacements in real environments, for the acquisition of relational spatial information, especially by children. However, a previous study using a virtual environment (VE) showed that children in a passive observation condition performed better than actives when asked to reconstruct in reality the environment explored virtually. Active children were unpractised in using the input device, which may have detracted from any active advantage, since input device operation may be regarded as a concurrent task, increasing cognitive load and spatial working memory demands. To examine this possibility, 7–8-year-old children in the present study were given 5 minutes of training with the joystick input device. When compared with passive participants for spatial learning, active participants gave a better performance than passives, placing objects significantly more accurately. The importance of interface training when using VEs for assessment and training was discussed.     

Is Route Learning More Than Serial Learning?

Abstract

The nature of route learning in terms of the memorizing of landmarks was investigated. In Experiment 1, participants memorized landmarks while being guided through a computer-simulated hallway (dynamic, with spatial context), or while viewing the landmarks one by one in front of a black background (static, without context). Two more conditions completed the 2 × 2 design. One condition preserved the dynamic landmark viewing properties (observers approached each object, passed it, turned to the next object, and so on), but the background was black (dynamic, without context). In the other condition the observer saw a stationary display of each object within a hallway, but did not approach the object (static, with context). Serial recall was much better after viewing the landmarks in the dynamic presentation format with spatial context than in the other conditions. Experiment 2 showed that the superior performance in the dynamic condition with context was abolished when all hallway segments were equally long. This implies that metric information is a component of route knowledge at a very early stage, which is incompatible with the dominant framework, but is compatible with the alternative framework for spatial microgenesis.     

无人机对移动目标干扰航路问题研究

针对无人机的作战特点,结合作战目标的运动特性,研究了无人机对移动目标干扰航路的问题,引入了垂直干扰航路和跟踪干扰航路的概念,并建立了相应的干扰航路模型.通过仿真对比验证了两类航路规划方法的优缺点,从而确保对移动目标干扰航路规划的有效性,为无人机的任务规划决策提供了有力的辅助与支持.     

基于分层策略的三维航迹快速规划方法

快速航迹规划能力是任务规划系统追求的目标之一。提出了一种基于分层策略的三维航迹快速规划方法。该方法分为两个层次：全局规划和局部规划。全局规划在综合利用战场信息的基础上,利用遗传算法规划出最优或次优的引导点集,该引导点列所在的区域为最优连通域,并且在该连通域内能找到可行航迹,全局规划利用引导点列大致指明了最优航迹的走向;局部规划根据全局规划提供的引导信息和战场信息,利用SAS(Sparse A Search)算法快速规划出满足攻击角度约束的平面可行航迹,高度规划采用速度较快的几何规划方法。仿真实验表明,该方法比SAS规划方法要快,且生成的三维航迹近似最优。     

Wayfinding Performance in and the Spatial Knowledge of a Color-coded Building for Adults and Children

The study reported here investigates the effect color has on the performance in a wayfinding task,the wayfinding strategies used and the acquisition of survey knowledge of a total of sixty participants: twenty second graders,sixth graders, and adults each. All participants had to find their way to a goal in a virtual environment with either a colored or a gray ground. After that,they had to find the shortest way from the start position to the goal in two consecutive trials,to draw a sketch map of the maze,and to mark the position of the goal figure in an overview of the maze. The results showed that color helped all participants to improve their performance in the wayfinding task, influenced the use of some wayfinding strategies but did not improve the acquisition of survey knowledge. Furthermore, there was a developmental achievement from second graders to adults. The results implicate firstly that children and adults similarly rely on the structuring of space through color when navigating through an unfamiliar large-scale environment,and secondly that there is a dissociation between wayfinding behavior and spatial knowledge.