icra13 paper comments-优快云博客

0 “ignored the number of game theory for auctions developed inthe last years” 这是最大的问题。

如果review comment这么讲，一般来讲说明在introduction里面这是没有说清楚的问题。而且是非常重要的问题。平时还是要多看paper,看相关paper引用的文献也是很好的方式。comments里面提供的几个citation：

Bertsekas, D.P. and D.A. Castanon, Parallel synchronous and asynchronous implementations of the auction algorithm, Parallel Computing 17 (1991) 707-732.

Zavlanos, M.M., Spesivtsev , L. and Pappas, G.J, A Distributed Auction Algorithm for the Assignment Problem, IEEE Conference on Decision and Control, 2008, pp.1212-1217.

1 “incorrect word usage” 非英语母语人员写英文这是时常的问题，需要以后更加注意

2 关于实验要“

analyze in different conditions”，这个在你在实际的机器人之后会更容易做。

现在你有virtualbox,你可以设置不同client可用的带宽，其实也是很好的方式。多试几组参数，就会有一定的统计结果。写在文章里面会是比较大的表格，会更有说服力。

3 有个review总结的很好，有些句子以后可以用上。明显是个native speaker写出来的:

This framework allows tomodel systems where the robots have incomplete informationabout the other robots. The proposed architecture is ahost-based network that consists of a data center, a cloudcluster host, and the robot clients. The auction chosen toactive resource allocation within this system is the secondprice auction. The host is the auctioneer that optimizesthe total revenue as well as the utility of every robotclient. The host also charges the clients for the usage ofthe network to maximize profit. The proposed utility forevery robot is strictly concave and increasing. A Nashequilibrium for this system exists, and the reader isreferenced to [27] for details.