But at some point, people have to be judged on their ability to code. Find me a better way to judge and I'll be all ears.
My C for Engineers class had a great model. Throughout the semester everyone coded with a common set of goals. Each assignment built on the previous one. The TA and prof were usually available either in the lab or their offices to answer questions. By the time the midterm and finals came around, most people had similar looking programs. Duh. We had common goals for our programs.
The midterm and final exams were interesting,though, and were true tests of our knowledge of what we had coded. We arrived at each of these exams, code on diskette, course notes in hand and Internet access disabled on each computer. Each person was given a slip of paper with a moderately difficult problem that had to be added to the existing code. In my case, it was to add some pretty gauges in ASCII text to my code. Those that had relied too much on others to get their code working, or who didn't have their code in proper working order, failed the exam. It was simple and fair.
Another point, is that your Lab isn't very good at programming/implementing bots. Assuming you're using a PIC microcontroller
Wow, that's a really nice assertion. I would like to point out that the RHex robot co-developed by our lab and a few other universities is arguably the most successful walking/running robot in the world. It is probably the most reliable ambulatory robot ever built and is currently undergoing field demonstrations at SwRI in Texas.
BTW, the robot uses a PC-104 stack running QNX. Yes, there's a PIC on it, incorporated on a custom PC-104 board, among all the other control electronics. It's not programmed with run-of-the-mill Radio-Shack "min/max control code" like your "basic claw-bot". Have a look at the papers written on RHex and the other robots and maybe you'll learn a thing or two about robotics.
Imagine: the onboard computer crashes and the knee motor decides to do a 360. Goodbye leg.
I've seen this happen on the legged robots here in the lab. When that happens we just hit the kill switch and resolder the broken wires. I'd hate to have the "exoskeleton" kill someone because of a computer hiccup.
The students are using university property to develop the code. They only pay a fraction of the cost via their tuition; the rest is obtained from other funding sources including public and corporate ones.
Since there are so many interests funding university resources, the answer as to who owns the intellectual property developed with these resources becomes a whole lot more difficult to figure out.
I've dealt with the University of Alberta's Industry Liason Office. Here is a summary of their performance since 1994. In the short term it is often diffcult to deal with them, but the overhead fees they charge are important: "The indirect costs -- or overhead -- of research include utilities (electricity, water, natural gas, and so on) physical plant (building maintenance, repairs), library support, financial and other administrative costs" (source). Check out that page since it'll even give you a percentage breakdown of where the overhead fees are allocated. For instance, 11% of the fees go to the university's libraries. That means that other students will get access to more books.
The ILO departments also provide important services to researchers such as patent background checks and market analysis. They're not just blood-suckers waiting to bounce on student- or professor-generated ideas.
Motion capture is not the final solution. Breaking down realistic human motion into mathematical equations is. Just ask the folks at Boston Dynamics. I saw a demonstration of one of their models (programmed by a graduate from my lab) and it was superb.
Pictures of Predator in Florida
on
Robots Go To War
·
· Score: 1
Check out a picture and another one a friend and I took of a Predator UAV in front of the University of Alberta's Polar Bear robot. It gives you a sense of scale. The Polar Bear is five feet long.
The pictures were taken at a robotics symposium in Orlando, Florida in July, 2000. Other pictures from the symposium with other UAVs as well as ground robot vehicles can be found here, here (includes rear and side views of Predator), and here.
Slashdot Mirror
I spotted two identical A/UX manuals at an IKEA. They were there about a month ago and were being used as props. Kinda neato.
But at some point, people have to be judged on their ability to code. Find me a better way to judge and I'll be all ears.
My C for Engineers class had a great model. Throughout the semester everyone coded with a common set of goals. Each assignment built on the previous one. The TA and prof were usually available either in the lab or their offices to answer questions. By the time the midterm and finals came around, most people had similar looking programs. Duh. We had common goals for our programs.
The midterm and final exams were interesting ,though, and were true tests of our knowledge of what we had coded. We arrived at each of these exams, code on diskette, course notes in hand and Internet access disabled on each computer. Each person was given a slip of paper with a moderately difficult problem that had to be added to the existing code. In my case, it was to add some pretty gauges in ASCII text to my code. Those that had relied too much on others to get their code working, or who didn't have their code in proper working order, failed the exam. It was simple and fair.
Wow, that's a really nice assertion. I would like to point out that the RHex robot co-developed by our lab and a few other universities is arguably the most successful walking/running robot in the world. It is probably the most reliable ambulatory robot ever built and is currently undergoing field demonstrations at SwRI in Texas.
BTW, the robot uses a PC-104 stack running QNX. Yes, there's a PIC on it, incorporated on a custom PC-104 board, among all the other control electronics. It's not programmed with run-of-the-mill Radio-Shack "min/max control code" like your "basic claw-bot". Have a look at the papers written on RHex and the other robots and maybe you'll learn a thing or two about robotics.
Imagine: the onboard computer crashes and the knee motor decides to do a 360. Goodbye leg.
I've seen this happen on the legged robots here in the lab. When that happens we just hit the kill switch and resolder the broken wires. I'd hate to have the "exoskeleton" kill someone because of a computer hiccup.
The students are using university property to develop the code. They only pay a fraction of the cost via their tuition; the rest is obtained from other funding sources including public and corporate ones.
Since there are so many interests funding university resources, the answer as to who owns the intellectual property developed with these resources becomes a whole lot more difficult to figure out.
I've dealt with the University of Alberta's Industry Liason Office. Here is a summary of their performance since 1994. In the short term it is often diffcult to deal with them, but the overhead fees they charge are important: "The indirect costs -- or overhead -- of research include utilities (electricity, water, natural gas, and so on) physical plant (building maintenance, repairs), library support, financial and other administrative costs" (source). Check out that page since it'll even give you a percentage breakdown of where the overhead fees are allocated. For instance, 11% of the fees go to the university's libraries. That means that other students will get access to more books.
The ILO departments also provide important services to researchers such as patent background checks and market analysis. They're not just blood-suckers waiting to bounce on student- or professor-generated ideas.
Motion capture is not the final solution. Breaking down realistic human motion into mathematical equations is. Just ask the folks at Boston Dynamics. I saw a demonstration of one of their models (programmed by a graduate from my lab) and it was superb.
Check out a picture and another one a friend and I took of a Predator UAV in front of the University of Alberta's Polar Bear robot. It gives you a sense of scale. The Polar Bear is five feet long.
The pictures were taken at a robotics symposium in Orlando, Florida in July, 2000. Other pictures from the symposium with other UAVs as well as ground robot vehicles can be found here, here (includes rear and side views of Predator), and here.