Closing Credits

Formal Acknowledgements

I would like to thank first and foremost, Prof. Naresh Jotwani, my guide and mentor, who not only gave guidance on algorithm design and implementation in this project, but was also the reason for its genesis. The discussions we’ve had on the subject matter have proved invaluable in determining the direction and scope of this body of work. Not only has he acted as a technical guide, but he also provided motivational “pep-talks” whenever required.

My sincere thanks to Akshat Kumar, Arpit Dhariwal, Prashant Tholia and Prasoon Gupta for letting me usurp their PCs for my work. A special mention for Abhishek Gupta who helped out with Bryce 5.

I would also like to thank the development team (Bloodshed) of Dev-C++ IDE and the Mingw compiler for creating a light, freeware environment for software development.

Conclusion

Lots of future work hopefully not including any bug fixes in my programs. The structure of my bot programs doesn't quite help anybody to set up a platform for further work, but it does give hints ;) Heres to a bot who can outthink a human given similar resources...

x1 AI (ugh!)

THE END

Penultimate

This will probably be the last update before the final presentation. D* was ditched because i had little time to understand it AND implement it. Used plain ol A* instead, first with recalculation on bumping into obstacle and then recalculation when obstacle comes into visibility range (still not sure if this is how it works ;) . Have a demo to give to Jotwani today, a report to rewrite tonight and print tomorrow morning, and last but not the least, make a presentation for the day after tomorrow.

Project D*

Demo uno on monday...have to implement and test D* before that. (?!?!)
jotwani sez take terrains of various roughness and show bot performance in the report. which means i gotta run a lot more tests now (which probably might fail).

Stentz's '94 paper on D* is what im using for reference. may god show mercy on me and my bot :S

Time to run for home base

Just finished 1st report soft copy. It has minimal experimental output. gotta change the crappy algo. will try and implement atleast A* if not D* although cant understand how situations goinna improve if i use the former.