| 查看: 1154 | 回复: 2 | ||
| 本帖产生 1 个 翻译EPI ，点击这里进行查看 | ||
[求助]
论文翻译。。英译中！！！翻译成中文。。
|
||
|
1、In order to make the humanoid robot walk freely in complicated circumstance，the reliable capabilities for obtaining plane information from its surroundings are demanded．A system for extracting planes from data taken by stereo vision was presented．After the depth image was obtained．the pixels of each line were scanned and split into straight line segments．The neighbouring relation of line segments was kept in link structure．The groups ofthree 1ine segments were selected as seed regions．A queue was maintained for storing seed regions，and then the plane region was expanded around the seed region．Th e process of region growing continued un til the queue of seed regions was empty．Atier trimming．the edges of the planes became smooth．In the end，cxtracted planes were obtained．In the experiment．two models were used：pipe and stairs．Two planes in pipe mode1 and six planes in stairs model were extracted exactly．The speed and precision of algorithm can satisfy the demands ofhumanoid robot’s navigation． 2、Humanoid robots were considered to have more advantages of moving in environments designed for real humans than traditiona1 wheeled robots．The humanoid robots had more degrees of freedom and could use more actions while walking．so their movement was more flexible．111ey could step on or over obstacles f 1]，climb up or down stair[2]，turn around and even crawl underneath the object [3],etc．In complicated environments． the foun dationa1 problem of walking freely for the humanoid robots was navigation．At present，some famous humanoid robot platforms such as ASIM0 of HONDA f41,WABIAN一2 of Waseda University,and HRP一3 of AIST have been developed for veri ing al1 kinds of methods of controlling and navigation．Although related techn ologies were reported，the humanoid robots were still premature to use in actual 1ife．and there were practica1 problems．One of them was the stability while walking[5—6]．The inherent instability of hum anoid robots should be considered firstly,which was the main problem to make it difi cult in practical applications． 3、The navigation system generated a path that a robot moved along from a current position to a target position based on the mapping information around the robot．In the previous navigation techniques，wheeled robots were studied[7～81]：however,there were few researches on the navigation system for the humanoid robots． The movement of a humanoid robot could be split into limited actions and each action a~anged deferent foot placements[91]．The deferent sequences of actions would lead to deferent paths．A hum anoid robot was equipped with two cameras as its eyes．These actions required reliable perceptual capabilities for obtaining its surrounding info rrnation．It was assumed that there were two kinds of oNects in the circumstance where the humanoid robots moved[1O]．One was the object having planar surface that the robot’s fo ot could put on，such as floor,low and flat object，and stairs．The other was the obstacle which the robot could not step over and should try its best to avoid．such as wall and large object．It was important to distinguish the objects with flat surface from obstacles for the safe navigation of a humanoid robot．To solve this problem，the plane must be extracted from surroun ding info rm ation．In this work．a system is presented for extracting planes from data taken by stereo vision 4、OKADA et al[11] proposed mat 3D Hough Transforiilation was used for candidate extraction from depth map information and the plane segment candidates were fired into the depth map in order to detect the partial plane segment．CHEN et al[1 2]presented a newmethod of segmenting planar regions when all uncalibrated camera underwent pure translation．Both methods could segment the plane from original images， but the information of plane in real worid would not be obtained．Therefore，these methods were not fit for the humanoid robots．The method in this work would not only extract the planes from images but also would get the information about the planes such as position，angle and height. |
回复此楼» 猜你喜欢
材料调剂
已经有10人回复
-
一志愿哈尔滨工业大学085600英一数二337分求调剂
已经有5人回复
-
288求调剂 一志愿哈工大 材料与化工
已经有11人回复
-
求调剂，一志愿南京航空航天大学 ，080500材料科学与工程学硕
已经有9人回复
-
本科211，293分请求调剂
已经有4人回复
-
求调剂
已经有3人回复
-
085400电子信息319求调剂（接受跨专业调剂）
已经有5人回复
-
268求调剂
已经有9人回复
-
一志愿双非085502，267分，过四级求调剂
已经有3人回复
-
一志愿085404，总分291，四级已过，求调剂
已经有4人回复
baoshanqiu
至尊木虫 (著名写手)
- 翻译EPI: 204
- 应助: 103 (高中生)
- 金币: 15293.5
- 红花: 41
- 帖子: 1611
- 在线: 431.6小时
- 虫号: 3856202
- 注册: 2015-05-08
- 性别: GG
- 专业: 胃肠道免疫相关疾病
【答案】应助回帖
★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★
123wy123: 金币+15, 翻译EPI+1, ★★★很有帮助, 非常感谢！！！谢谢 2015-06-22 11:50:12
123wy123: 金币+15, 翻译EPI+1, ★★★很有帮助, 非常感谢！！！谢谢 2015-06-22 11:50:12
|
1.为了让人形机器人能在复杂的环境自由行走，需要可靠的从周围获得平面信息的能力。本文报告了一种通过立体视觉系统得到的资料提取平面的系统。得到深度图像后，每一条线的像素经扫描被分成直线段。线段的相邻关系被保留在联结结构中。由三条线段构成的组被选作种子区域。将存储的种子区域排队后，平面区域就围绕种子区域扩张。区域生长的过程持续进行到排队的种子区域用完为止。经过修饰，平面的边缘变得光滑。最后得到提取的平面。试验中采用了两个模型：管道模型和楼梯模型。确切地有两个平面用的是管道模型，六个平面用的是楼梯模型提取。算法的速度和精度能满足人形机器人导航的要求。 2.人形机器人较之传统的轮式机器人在为真人设计的环境中运动有更多优点。人形机器人在行走时有更多的自由度，能采取更多的行动，因而它们的行动更灵活。它们可以踏上或越过障碍物 [1]、上下楼梯[2]、转过身来甚至从物体下面爬过 [3]等。在复杂的环境中，人形机器人自由行走的基本问题是导航。现在，一些著名的人形机器人平台如本田 f41的ASIM0、早稻田大学的WABIAN一2和 AIST（国家高等工业科学技术研究院）的HRP一3都采用了各种控制和导航的方法。尽管有相关的技术报道，人形机器人还不能用于现实生活。仍存在一些实际问题。其中之一是行走时的稳定性[5—6]。首先应考虑人形机器人固有的不稳定性，这是它难以实际应用的主要问题。 3.（待续） |
2楼2015-06-19 05:05:27
baoshanqiu
至尊木虫 (著名写手)
- 翻译EPI: 204
- 应助: 103 (高中生)
- 金币: 15293.5
- 红花: 41
- 帖子: 1611
- 在线: 431.6小时
- 虫号: 3856202
- 注册: 2015-05-08
- 性别: GG
- 专业: 胃肠道免疫相关疾病
【答案】应助回帖
★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★
123wy123: 金币+15, ★★★很有帮助, 非常感谢！！！谢谢 2015-06-22 11:50:22
123wy123: 金币+15, ★★★很有帮助, 非常感谢！！！谢谢 2015-06-22 11:50:22
|
3.导航系统生成一条路径，机器人则沿着这条路径根据周围的映像信息从当前的位置移动到目标位置。过去的导航技术研究了轮式机器人[7～81]，但很少有导航系统涉及人形机器人。人形机器人的移动可以分割成有限的动作。每一个动作将脚放在不同的位置[91]。动作的顺序不同会产生不同的路径。人形机器人装配有两台摄像机作为眼睛。这些动作需要有可靠的感知能力以获得其周围的信息。设想人形机器人移动的环境有两类物体[10]。一类是机器人可以涉足的具有平坦表面的物体，比如楼板，矮和平的物体和楼梯。另一类是机器人不能涉足的障碍物，应尽可能避开它。比如墙和大的物体。为了人形机器人安全导航，重要的是区分表面平坦的物体和障碍物。要解决这个问题，必须从周围信息中提取平面。本研究介绍了一个从立体视觉资料中提取平面的系统。 4.OKADA 等[11]提出用mat 3D Hough Transformation作为从深度映像信息提取的候选，为探测部分平面节段而将候选平面节段引入深度映像。CHEN等[1 2]报告了一种当所有未经校准的摄像机纯直译时分割平面区域的新方法。两种方法都能从原始图像分割平面，但无法获得真实世界的平面信息。因此，这些方法不适用于人形机器人。本文的方法不仅从图像提取平面，还能得到有关平面的诸如位置、角度和高度的信息。 |
3楼2015-06-19 23:04:44
