본문 바로가기

Thoughts/Technical Writing

Robotics paper


Nice expressions


Introduction


[1]의 첫문단 (로봇 어플리케이션에는 여러 제약이 있고, 실제 사용할 때는 적은 수의 set 중에 좋은 것을 고른다. 비행기 같은 것은 이러한 문제에 특히 더 위험(acute)하다.)

Mobile robot applications most often impose time and resource constraints on their motion planning software. Real robotic system replan at a fast rate in order to incorporate a constant stream of new perception data. Due to time constraints, most planners operate by considering a relatively small set of possible actions and and commanding the best choice before the next execution deadline arrives. 

Examples of such systems include ground vehicles (driving on the highway at high speed or navigating through cluttered, off-road terrain), unmanned ariel vehicles including rotary and fixed-wing craft. 

Failure to generate a plan quickly and with sufficient lookahead risks damage to the vehicle. This concern is particularly acute in the case of fixed-wing aircraft, which do not possess the capability to stop in place if the planner misses its deadline. 

[2]의 첫문단 (path set이 무엇인지. 이 논문에서 하려는 것)

A path set consists of a fixed set of control sequences, each paired with its corresponding trajectory. These controls are often selected, and their responses precomputed, offline using a high-fidelity vehicle model. Since the paths are represented by actions, they are inherently constructed in the vehicle's coordinate frame and originate from the robot's current position. The salient feature of a path set considered in this paper is the spatial separation, of path diversity, of its constituent paths. This paper examines the relationship between path diversity and planner success rate. 


  

[1]의 두번째 문단 (이를 푸는 기존의 방법은 set을 precompute하는 것이다. 장점과 단점이 있다.)

One popular strategy for improving online planning time is to precompute a set of candidate actions a priori [a,b,c,d]. The advantage of this is that no online computation time is wasted in considering paths that robot cannot execute. The drawback is that a path set generated offline may not incorporate any knowledge of particular obstacle configurations. 

[2]의 두 번째 문단 (path set이 무엇인지) 

Path sets are the basis of an approach to planning and control suitable for use in real-world planning problems. As an alternative to planning in the continuum of motions, a path set considers a discrete set of control inputs, picking the best one according to its predicted trajectory. The planner selects one path for execution each cycle, but it may switch paths with each consecutive cycle, at a rate of roughly 10 Hz. 



[1]의 제안 알고리즘 문단 (우리가 제안하는 것은 주어진 path set을 evaluate하는 것이다. 이를 위해서 이것저것 제안)

Our work discusses the problem of evaluating a path or set of paths in the context of unknown obstacle locations and density in the world. Two theoretical algorithms for computing an exact probability of collision are derived. Two approximate algorithms (A and B) are then presented, which provide a tractable solution to the problem of selecting a diver set of paths. These algorithms cull a large path set down to a smaller, more manageable size, while attempting to maximize the robustness of the path set with respect to unknown obstacle configurations; see Figure 1. The choice of path set size for a particular application may depend on constraints such as planner requirements, available memory, and CPU performance. The reduced path sets are analyzed with respect to their survivability in the presence of obstacles. 


[2]의 Related works 설명 

The concept of diversity has only recently been appreciated as an aspect of the path set planning problem that can make or break the planner on challenging problems. But what is path diversity? 

Green and Kelly define relative completeness, as "the prior probability, before the environment is specified, of producing a solution path in a bounded amount of computation". 




[a] E. Frazzoli, M. Dahleh, and E. Feron, “Real-time motion planning for agile autonomous vehicles,” AIAA Journal of Guidance, Control, and Dynamics, vol. 25, no. 1, pp. 116–129, 2002.

[b] J. Go, T. Vu, and J. Kuffner, “Autonomous behaviors for inter- active vehicle animations,” Int. J. Graphical Models, 2005.

[c] M. Lau and J. Kuffner, “Precomputed search trees: Planning for interactive goal-driven animation,” in Proc. ACM SIGGRAPH / Eurographics Symposium on Computer Animation, 2006

[d] S. Thrun et al., “Stanley: The robot that won the DARPA grand challenge,” J. Field Robotics, vol. 23, no. 9, pp. 661–692, 2006.


'Thoughts > Technical Writing' 카테고리의 다른 글

Reviews I got from RSS 2017  (0) 2017.05.10
Another Reject from ICRA 2017  (0) 2017.01.26
호프스테더  (1) 2014.11.10
노인과 로봇  (3) 2014.03.13
Proposal for Mobile Computing  (0) 2013.10.27