Submitted Questions and Answers

Mail new questions for the Competition to Prof. Nicholas Krouglicof, krouglin@union.edu

We have a question about four pneumatic cylinders that might cause a clearance issue in the tripwire competition. These parts are fully adjustable and/or removable as they are not required for the event. I was unsure if this is legal though. Any clarification would be helpful. As a general rule, manual configuration changes are not allowed. Keep in mind the basic premise of the competition. The competition is supposed to simulate an unmanned mission to Mars therefore there isn't anyone available to make an adjustment. Your robot should basically be configured for all events at all times. Just a few quick questions, we are planning on using a camera system that hooks up to the RCA on a TV. Is there a TV with RCA jacks that we could use at competition? Also, for the technical presentations is there a going to be a projector/screen for power point? There will be a video projector and screen available for power point. As a general rule, we are assuming that the participants will supply their own laptops. As per your request for a TV, I will forward your request to Media Services at Union College and get back to you. Hello, I'm a student here in Mexico, and my question is the next, I have several doubts, about the process to introduce the robot to your country, specifically, I've been told we need a signed letter by the authorities of the WMC, in UNION COLLEGE, so it can be authorized the entry of the robot to the US. If you can please, help me with this problem, and give some kind of guidance, I will be truly thankful, and in case we really need this letter, could you please send it or in case you're not the person I need to approach to, could you please send me the e-mail address. I will send you a letter. If I send you a signed letter as a PDF file, would that be satisfactory? Let me know. [NOTE: Please contact Prof. Nicholas Krouglicof if your team needs any kind of documentation of this type.] You've had contact with our faculty advisor about our concern with the re-filling of our air tanks. The information we got was that there is a divers store 20 min away from the college where we can fill them up. We believe that's quite inconvenient since the time available between first attempt and second attempt for each trial is not much and it would take us approximatelly from 50 min to one hour to drive there, fill our 5 tanks and be back. The solution that was presented last year was that one of this divers stores nearby brought the equipment required to offer the filling service to all teams, inside the campus, charging a reasonable fee for each refill. We are really concerned about this. We believe that it could represent a problem to the challenge considering that in previous events most of the robots had been pneumatic powered. He have considered your suggestion about refilling the scuba tanks on campus. Unfortunately, this will not be possible. The College as well as the diving store have concerns regarding liability (i.e. responsibility in case of an accident). In fact due to concerns over liability, the diving store will only fill tanks for licensed divers. As I mentioned in my previous E-mail we will have to provide a letter stipulating that the tanks will not be used for scuba diving. One possible solution is to rent enough tanks to compete all six events. Several diving stores offer equipment rental. We've been making the arrangements to transport the robots to the competition. A delivery company is guiding us through the process, and has told us that we need certain documents and information from the host university. The following information is needed: 1. complete address of the university and department that is going to take care of receiving the robots 2. A copy of the ID of the person responsible for such operation One more thing we need to know is if this department in charge of receiving the packages has the necessary equipment to load and unload the robots from the delivery truck. I'd appreciate if you could send us this information as soon as possible in order to continue with the rest of the process. Any question you have, I^Òll be glad to answer. You walking machine should be shipped to: Union College att: Mr. Roland Pierson Engineering Machine Lab Schenectady, N.Y. 12308 USA Mr. Roland Pierson is the Director of the engineering machine shop. He has agreed to supervise the unloading of all the walking machines and to store them in the machine shop until your arrival. He has a forklift at his disposal therefore unloading should not be a problem. [Note: This question has been slighted edited from the original submission.] I actually was aware of both the April 13 deadline and the ten working days deadline. I had chosen to ignore the April 13 posting simply because the ten working days requirement seemed more standard and because, quite frankly, it suited the students better. .... I would appreciate it if you allowed us the ten working day deadline, April 16. I'm going to have to be firm on the deadline for the submission of the technical papers. In all fairness we would have to offer the same extension to all the teams participating in the competition and given the tight schedule this is not an option. The team of judges who will be reviewing the technical papers are already scheduled to meet on Wednesday, April 14 to begin the review process. Several of the judges are from local industry (GE for example) and they have already planned for this date. To ensure that all the papers are reviewed fairly, we are planning a marathon judging session involving all the judges in the same room reviewing all the papers at the same time. I hope you understand the situation and I encourage you to try to meet the deadline of April 13. I will describe possible control paradigms for our robot. Can you please point out where they violate Autonomous Level 2? Suppose we use 3 beacons. The beacons use ultrasonic sensors to determine the range to the robot. These ranges are sent to the robot over a wireless link. The robot uses the 3 ranges to determine its own location. The beacons use the 3 ranges to determine the robot's location. The robot's location is transmitted to the Robot over a wireless link. The beacons get reflections from objects on the course. The beacons send the range to objects on the course to the robot wirelessly. In general, for autonomous level I or II the robot must be completely self-contained, i.e., the power source and control computer must be onboard. For this reason, I am somewhat concerned about performing too much computing in the beacons. Basically, I believe that any of the scenarios that you describe are acceptable providing that the beacons are presented (in your technical paper for example) as "intelligent" sensors. In an "intelligent" sensor, a processor is typically used to control and process data from a transducer and to communicate the results to a host. As long as the processors located in the beacons are not directly  involved in the control of the robot, I think that all the scenarios are acceptable. NOTE: a robot that relies on beacons for navigation will be scored as autonomous 2. If a team announces their intent to run autonomous 2, the judges will intentionally misalign the robot in the starting box. If the robot automatically corrects its trajectory, then the run qualifies as autonomous 2. The only exception is Event 5 "Object Seeking". In event 5, the test for autonomous 2 is whether or not teams use the coordinate information provided by the judges. A random search pattern that "sweeps" boxes 4, 5, and 6 is also autonomous 1. To achieve autonomous 2, the team must rely on sensor information to actively locate the block within boxes 4, 5, or 6. The use of beacons positioned at the three predetermined locations is still allowed for autonomous 2 providing: the team does not rely on a random search pattern the team does not use the block coordinate information the team does not place a beacon on the block Our walking machines are of the pneumatic type, and the specifications for the air tanks are as follows: SPIRO breathing tanks, model 9030 for 4500 psi. We would like to know whether or not the Union College will provide for the tank refilling service.  In case it will, what would the cost be?  Otherwise, is there a nearby place where tanks can be refilled? I've located a diving store that can most probably fill your tanks. They are located approximately 20 minutes from Union College. You can check out their WEB site at: http://www.capitalandscuba.com Their air bank goes up to a maximum of 4500 psi so they may not be able to charge your tanks completely. In addition, they need to see the exact charging valve on your tanks before they can provide a definitive answer but they are fairly confident. Unless you have a licensed diver on you team, they will need a letter from the college stating that the tanks will not be used for diving purposes. We are currently completing our technical paper for the upcoming competition and we had a question concerning the exact contents of this document. Specifically, should we include manufacturing drawings, electrical schematics and technical specifications on the different components used? I would say that none of the items listed above should be included in the main body of the technical paper. At your discretion, some detailed drawings may be included in the appendix if they describe critical components that are specifically described in the paper. Otherwise, I would limit the drawings in the technical paper to assembly drawings, exploded views, block diagrams, flowcharts, etc. I would, however, suggest that you have all the detailed drawings and technical specifications available for the judges during the competition. This is particularly true for components that influence the safety of your design. We have a question concerning the "E-STOP". For reasons of accessibility, we are considering the possibility of installing TWO emergency stop buttons in series. We realize that the rules call for a single emergency stop that cuts the power to the robot drives. Are two buttons in series acceptable? With this configuration, if either button is depressed, the power to the drives is cut. Note that a relay is absolutely necessary since no emergency stop button can realistically withstand a current of 50 A DC. In my opinion the two E-STOP buttons in series is an acceptable configuration if accessibility is a problem. Note that the buttons must not require sustained action to operate and that the relay must be "normally opened". I am uncertain about the exact deadline for the technical papers and the precise mail account to send them. Also I would like to know if there is a webpage or link to get the current format for the papers, (I have problems with SAE´s page links). And finally, if my team is interested with the "online publication" , do we need to fill any special form ?? Please explain more about waiver/release/authorization...is it a sepecial file we need ?? ,how do we sign it..?? The exact deadline for submitting your technical paper can de found on page 43 of the rules: * Technical Paper submission deadline - Midnight Eastern Standard Time, Tuesday, April 13, 2004. * Submit technical papers (exclusively) to: robot@engineering.union.edu The instructions for writing an SAE Technical Paper were moved during the reorganization of the SAE WEB site. They can now be found at: http://volunteers.sae.org/authors/present.htm The waiver for on-line journal publication was inadvertently left out of the rules package. I will forward a copy by E-mail to all participants shortly. I have noticed a few discrepancies in the dates listed for the conference. The official SAE website lists the dates of the conference as 04/28/2004 - 05/01/2004, the front page of your website lists the dates as April 29th through May 1st, and the challenge schedule only lists events to be held on Friday, April 30th and Saturday, May 1st. Which of these sets of dates should we plan on? Also, what is the final deadline for registration? The competition runs from Thursday, April 29 until Saturday, May 1. Thursday is reserved for on-site registration and setup in the Field House. The competition really starts on Friday with the technical presentations, safety inspection, and the first two events (Dash and Load Retrieval). The remaining events are on Saturday as is the awards banquet. If possible, you should plan on arriving sometime on Thursday to register your team with the SAE, sign the necessary insurance forms, and setup your pit area. You can consult the Union Web page for more information: http://engineering.union.edu/WalkingMachine/schedule.html The final date for registering a team is March 5, 2004. I have one more question regarding Pneumatics. Will there be a filling station there at Union? And, if not, where can we fill our tank? And then what type of valve do we need so we can fill it? Just want to make sure that we are prepared for the competition. Send me the specifications on your tank including the type (i.e. scuba, fireman, paint-ball, etc.) and I'll check if we can refill it on campus. If we can not, I'll check for nearby facilities that can help you. [NOTE: this brings up the general question of "special needs". Please think about your project's "special needs" and prepare for them in advance. Let us know about any such need and we will let you know to what extent we can help.] We have a dynamically stable robot (six leg, but running dynamically... it doesn't need control to stay up, but it may flip over when climbing obstacles). Our robot is pneumatic. In the rule, it's specifed that I must use a manual dump valve... The problem is that with a running robot, it's really difficult to reach this valve when the robot is moving. The only solution that I found to remotely operate a dump valve is to use a solenoid valve, activated by a normaly open relay drived by a kill-switch (so if there is a problem with the switch, wire or power the robot will stop moving), to remove pressure from all components of my robot. Is this system allowed by the rules? I notice a line in the rule where it's specified that using a solenoid valve to drain pressure from the system is not allowed... can you tell me why? If this is not ok, can you suggest me another way to remotely operate a manual dump valve? The rules clearly state that "Using a solenoid valve to drain pressure from the system" in a pneumatic robot is unacceptable. In the configuration you describe the solenoid valve (which is already problematic) is driven by a relay which could also fail. Safety is of paramount importance in the competition so I would say that its better to err on the side of caution. I suggest a manually operated dump valve on the robot that satisfies the rules in addition to a remotely operated solenoid valve that de-pressurizes the system in the event of a problem. I believe the judges would appreciate the fact that you made every effort to ensure safety. We are in the process of hooking up our Pneumatics and had a question on Rule 22.5.5 regarding the E-stop for Pneumatics. Right now our valves are normally closed meaning if the E-stop is activated pressure is removed from the system however the actuators are locked in place. This means that they can not move but will hold the weight of the robot. Is this design sufficent for rule 22.5.5? According to Rule 22.5.5 "The E-Stop function for pressure systems should remove pressure from all components except the pressure tank itself." My understanding is that your E-Stop does remove the pressure source from the valves that control the cylinders but that the cylinders remain pressurized so as to support the weight of the machine. In this case, you should address 22.5.5 (b) Required Analysis (i.e. can each component that is pressurized following actuation of the E-Stop withstand a substantial pressure increase?). In general the primary concern of the judges is safety. Does your design pose any safety concerns for the operators, spectators or judges? For example, one potential safety concern that you might need to address is pinch points. If, for example, a team member gets his or her hand caught in the leg mechanism, how do you release the appropriate cylinder following actuation of the E-Stop so that the individual can free himself or herself? In summary and based on the information provided, I believe that your E-Stop implementation is adequate providing there are no additional safety concerns associated with your design. I was wondering if the dimensions, color, and shape were available for various objects used in the competition. The objects I am concerned with are the flags and poles used in the slalom event, the tires, the lumber, and the trip wires. Any information you can provide me will be greatly appreciated. The primary objective of Event 6 is to test the endurance and maneuverability of the walking robots in an unstructured environment. Consequently, the precise layout of the course including the size and shape of the obstacles is left (intentionally) unknown to the participants prior to the event. Typically the poles for the slalom event are 1/2" wooden dowels that are 1 meter long. With respect to the trip wire we plan on using a wire with a relatively high contrast relative to the background (probably black). We will make more information available to all the participants over the coming weeks via the competition web site. Images of the competition venue (the Field House) will be posted shortly. I have a question on the rules for this years Walking Machine Competition. This question refers to the emergency stop rules. We have a PC104, which acts as the "brain" of our robot and runs off of a 9V battery, that sends signals to our controller and eventually to our motor drives which run off of our main power source of six 12 V batteries. Am I correct that we must have one E-Stop button that disconnects both the 9 V battery (to the PC104) and the 12V batteries (of the main power source)? Or is it that the E-Stop only needs to disconnect the 12V batteries that powers the motors? As stated in the rules, the robot must have a single emergency (E-stop) that "fully and irrecoverably disconnects the power source from all drive systems." If the on-board computer is powered by a second battery, then you are not required to disconnect the battery from the computer by means of an E-stop. Be sure to adequately document your design so that you can prove to the judges that all "motive components" are disconnected from the power source when the E-stop is activated. I am writing about the E-stop rules for the SAE Walking Machine Challenge. Our robot will be able to flip over, therefore will have no definite top or bottom. Do we need to have E-stop on both sides? It is actively stable, and according to the rules it needs a remote E-stop. Does this mean that we need three E-stops? What I read in the rules is that the robot needs to have a single E-stop, but I am uncertain due to our complications. Can you please clarify these aspects for me? You've raised an interesting issue that we hadn't considered when drafting the rules.  So to answer your question I'll start from the objective of the E-Stop which is to enable the robot to be safely, quickly and absolutely stopped at any time by either the operator or a judge.  Or to put it simply, we always want to be able to "turn it off".   If a robot operates in different spatial orientations it is still necessary to have some type of absolute stop system that always accessible.  Since most robots have an obvious top and bottom rule 22.5 "Emergency Stop" was drafted with that type of configuration in mind.   Since you are designing a robot without an obvious top and bottom a single, fixed, E-stop would not work since, in some instances, it could concievably be under the robot.   What I'm going to propose is that we suspend the portion of rule 22.5.1 requiring a "single emergency stop" and rule 22.5.2 (1) "on the top of the robot"  and leave it to you to design a system that enables the robot to be easily and completely stopped by the operator or a judge no matter what its orientation.  All the remaining portions of rule 22.5 would remain in effect. As to the remote stop the the goal remains the same - to be able to safely turn the robot off at all times.  Is an additional remote stop necessary for your robot to meet that goal?  Without knowing a lot more about your design it's impossible for us to answer that question.  But, knowing the objective of the E-stop, you can determine the answer. In our design, we have decided to couple 3 joints on the robot to be driven by one motor for ease of turning.  I am not sure whether or not this design is legal.  I would appreciate it if you would review it and let us know how we should proceed. After reviewing your design concept, I'm assuming that three of the six legs on your robot are fixed and consequently there is no relative motion between these three legs and the frame (or for that matter between these three legs and each other). Strictly speaking this would seem to violate the rules, however you could argue that the three fixed legs are part of the frame and that the robots actually rests on the frame between steps. This would then be legal. There are also past precedents for such a design. Many teams - particularly rookie teams - design "beam walkers" which resemble two four-legged tables sliding over one another. Again there is no relative motion between the four legs of each table but you can argue that the table is actually a single leg. Beam walkers have always caused controversy but the judges generally allow them. I would also remind you that all robots must be capable (at least in theory) of completing all events. While a failure to convince the judges that your design can complete all events does not lead to disqualification, it does significantly reduce your score in the design judging. I'm not sure whether or not your robot can get over the trip wire. In short I would say that your design appears legal based on past competitions. What are the criteria for entering two robots from one institution? For example, is the $500 entry fee required per robot or per institution? Registration fees at all SAE Collegiate Design competitions are per entry, not per university. If you are entering 2 robots they will have to be registered separately. This year we are developing a vision system for a previously built robot. We are, however, unaware of the stipulations on using flags or beacons in Autonomy Level 2; the rules seem unclear. Are there any events where our score will be lessened by using a beacon to triangulate our position? In general, a robot that relies on beacons for navigation will be scored as autonomous 2. If a team announces their intent to run autonomous 2, the judges will intentionally misalign the robot in the starting box. If the robot automatically corrects its trajectory, then the run qualifies as autonomous 2. The only exception is Event 5 "Object Seeking". In event 5, the test for autonomous 2 is whether or not teams use the coordinate information provided by 1. A random search pattern that "sweeps" boxes 4, 5, and 6 is also autonomous 1. To achieve autonomous 2, the team must rely on sensor information to actively locate the block within boxes 4, 5, or 6. The use of beacons positioned at the three predetermined locations is still allowed for autonomous 2 providing: o the team does not rely on a random search pattern o the team does not use the block coordinate information o the team does not place a beacon on the block So if its your intention (and I'm guessing this is what you have in mind) to use three beacons in conjunction with a vision system in event 5, then that would qualify as autonomous 2. In Autonomy Level 2, is the robot required to take more than one positional fix or is a one time fix sufficient for this level? If you choose to run an event in autonomy level 2, the judges will intentionally misalign your robot in the starting box. For example, the robot may be positioned in one corner of the starting box with an initial orientation of 35 degrees relative to the center line of the course. If the robot automatically compensates for this initial misalignment, then this is autonomy level 2. If your robot only requires one positional fix, then that is sufficient. In general, the initial misalignment will be severe enough to ensure that any preprogrammed trajectory brings the robot out of bounds.