Maybe this was your plan already: Wouldn’t it be easier to use springs or rubber bands to keep the gripper in a default hold position. The controller would then only be used to open it and control grip pressure. You could also change the bands to adjust the max pressure “grip” for a given task. As far as getting a gripper material in (per previous post by Mccullermi) you could cut slots and use off the shelf o-rings. Oh one other thing, if you use a default grip mode, you will need some kind of back-up release so the ROV doesn’t get planted at the sea bottom if it grabs something too big to carry.
Found this while searching for ideas last night:
This one might be adaptable to a use an open frame outrunner motor:
@mccullermi That’s a great idea! I can easily incorporate a ribbed tip into the CAD file and @hogwildmy I’ll look into O-rings. My other thought was inlaying wire into the gripper so that the rubber jacket is used as a similar function to the O-ring, but I have not personally tested either of these approaches yet. My design challenge is to reduce the number of parts in the gripper to a few as possible. This relates to the O-rings because I would most likely sandwich the gripper material between a piece of acrylic. This would mean that I would have to have another piece of acrylic above and below the gripper material creating a total of three pieces of acrylic total per gripper finger. To avoid this design I still need to create an easy way to mount the gripper material. A project for this week as well as others.
@hogwildmy Making the gripper auto return or auto open actually kinda fits in with the current design for I am currently experimenting with putting a second joint in each finger which is held in position until closed with a torsion spring. I could theoretically just add more springs to the base of the gripper. But to be honest, I have actually not worked out the drive mechanism from the servo to the gripper yet.
Looks like I have my work cutout for me!
Version 2.8 Announcement
Perhaps you could make one of the claws with some grippy material, and the other 2 just as you have them.
Something you may want to consider when designing the jaws are some real world scenarios. Here’s one: I recently used my DTG2 to recover some plastic that had been discarded into the harbor by a work crew. The item was about 8 feet long, weighed about 15 lbs (in the air) and immediately sank nearly entirely into the soft, silty bottom. Had I not been able to tilt the jaws down (the DTG2 has 3 DOF) and sink the bottom jaw into the soft, silty floor, I wouldn’t have been able to recover it.
SCORE! Pre-Order an Underwater Servo Prototype
Just to give you a brief update, we made the first working prototype of the gripper today and mounted it on an ROV.
You can find a video of us using it in the test tank here ( https://www.youtube.com/watch?v=D1FtwtwrFcI )
Additionally, the waterproof servo is now for sale in the OpenROV store which can be found here.
( http://store.openrov.com/collections/experimental/products/prototype-underwater-servo )
<imgsrc="/uploads/openrov/original/2X/f/ff84e38f422505c04cfb756eee3d2c465237f422.jpg" width=“690” height=“388”>
Looks great, and I’m glad it works well!
I know I’m going to get grief for saying this, but that manipulator looks -real- wimpy. Besides perhaps a MATE competition, I can’t think of too many uses for something this weak.
Well, the servo itself is only rated at 2.8 to 3.4 kg/cm of torque. It has some power but won’t be doing any heavy lifting.
Also, the OpenROV model itself is pretty much balanced on a knife-edge in terms of buoyancy, so it won’t be picking up anything heavy anyway. You saw the struggle in the video. I think it’s good for adding a capability (grabbing) where there was none while still remaining somewhat affordable (the servo at $125 is going to keep this out of reach for a bit).
I know it looks weak as Stretch pointed out, but I can think of a number of potential applications, first off being able to pick up a gopro in the video is pretty cool. I could see this being used to retrieve even large items by being able to carry a loop or hook attached to a lift line down to an object and either hooking it or setting the loop and then releasing it to be retrieved with the haul line once the ROV is out of the way. I have been hoping to get a claw setup to be able to connect haul lines to old crab pots that are ghost fishing and this might be just the ticket.
The utility of this particular gripper may appear, at first, somewhat limited as it does not have a lot of power to go down, grab hold of an object and bring it back to the surface. However, I am inclined to look at it in a slightly different way.
One of the most important jobs a small observation class ROV can do is assist in the search/location of submerged objects. Police, and rescue organizations use small ROV’s for this purpose all the time. In this mode of operation it is not necessarily important to retrieve found objects, but rather to locate and mark (tag) found objects for investigation and later possible retrieval by divers.
In the search and location mode, a small gripper is really all that is needed. Simply attach the tagging clamp, beacon or weighted end of the marker buoy line to the gripper and release the marker at the location of the found item. In the case of a clamp, the gripper would be used to hold the spring loaded clamp open until the clamp is placed on to the found item.
If all you want to do is send a tag line down, then you could probably make something out of a carabiner and some velcro that would work (there was a guy who did just that a few years ago, I’ll post a picture if I can find one).
@Kevin_K - You made some good points about the overall power and load carrying ability of the OpenROV. It’s just not suitable for anything but the lightest of tasks. But on the off chance that someone is using OpenROV electronics with more powerful thrusters, etc., a grabber made from anodized aluminum would likely serve more purposes. Include a “locking device” that’s triggered by your servo and is able to hold as much weight as your tether can handle.
Great work, Patrick! I currently work with Work Class ROVs in the oil industry and from my short experience in the field what I take is “keep it simple”. I have worked with some manipulators and the one that we could always trust is the Atlas’ design (http://www.hydro-international.com/wosimages/1013_345.jpg), but it does need the wrist rotation system to make it really functional.
Nonetheless, I’ve been into the study of shape memory materials as actuators. They are cheap and allow a variety of design setups. If you want to expand your line of thought you should take a look at that. The power drain could be a problem though.
Addressing some of the concerns above, The gripper arms themselves have recently been bumped up to 6mm acrylic to provide more surface area and to make the device sturdier and less “wimpy” and although small it does provide a surprisingly good hold on things mainly because the servo is metal geared. As seen in the video it was able to hold onto the GoPro with a marginal grip.
Backing up, the goal of this design was to make a general gripper that could work with the waterproof servo and could adapt to as many situations as possible. Starting out I fully acknowledged that this gripper was not going to be perfect for every application, but would instead be a good general tool. Given that info, and keeping in mind the limitations of the power and size of the ROV
If we were to produce another type of gripper what type would you find the most useful for what you are doing? The next design I was thinking about would most likely involve some type of scooper.
I can’t seem to find much on this subject on google. Can you post a link to an actuator that uses this design?
A grabber that could be locked closed so you could recover fairly heavy items (up to the breaking strength of the tether.
A grabber that could be used to grab / collect soft objects (such as sea cumbers, etc.).
A suction grabber (for small and/or soft sample collection)
A rope cutter (to clear ropes, kelp, etc. from a boat prop, etc.)
I am working on an assingment about robot grippers and I am searching for the influence due to the amount of its fingers. Do you have some source about it?
Nayara S. M.
I don’t have any specific articles, but Yale published their academic results here: http://www.eng.yale.edu/grablab/pubs/Dollar_IJRR2010.pdf
There are some finger comparisons.