I was hoping to generate interest in developing a Digital Human for ergonomics assessments in Blender.
According to the ILO, the annual cost to the global economy is a staggering 1.25 thousand billion (trillion). That number doesn’t account for the human misery associated with workplace injuries. Unfortunately, commercially available digital human models for addressing ergonomics issues are both expensive and difficult to use.
In this case, the distinction is that this human model would need to be scalable using various anthropometric databases, have accurate posture prediction and enable developers to create risk analysis tools by reading the dof and joint locations for the manikin. A typical example of ergonomic analysis would be Rapid Upper Limb Assessment (RULA) tool.
Is there any interest? It’s a lot to digest, but would impact a lot of lives.
cad would be better than blender. mesh’s are hollow while cad objects are solids with varying densities. thats why cadd is used to prototype military wepons, do fire assesments, structrual damage predictions etc… you could predict where a human mesh was going to get hit and maybe even the enery force of the impact, but you couldn’t get the internal damage.
We’re talking about different classes of human models. In your case, the model would be intended to measure acute trauma to the body such as the ones used in vehicle crash simulation.
In this case, we’re really looking into identifying ergonomics risk by accounting for factors like posture and link length.
perhaps i dont under stand. if all you want is human meshes of diferent sizes, shapes, genders, etc… use make human. no need to reinvent the wheel. http://makehuman.blogspot.com/
I agree, if you need human male or female models that can be tailored to your wishes in term of overall shape/size/weight/etc… makehuman is a great choice, and as it does not even require any kind of modelling knowledge, it is very user friendly.
After that, you can always export your makehuman tailored model into Blender if you need specific editing, Makehuman has some very nice scripts playing very well with Blender with an already-made armature.
Definately makehuman models imported into Blender would be the favourable choice. Once rigged with an armature (skeleton), the armature could be scaled with various proportions to deform the mesh according to “anthropometric databases” :eek:
The joints in the armature can have rotation limits set on them, restricting the degrees of freedom.
The armature can then be posed and the joint orientations compared mathematically to some ideal (or set of ideals).
All this is possible in Blender using models generated by makehuman, is this what you were talking about?
The hardest parts are the construction of the armature to acceptable standards (for which you would need to explain your precise requirements to a skilled rigger) and the mathematical analysis of the joints (for which you would need to explain your precise requirements to a scripter in Blender or a programmer for a stand alone application)
Is it possible to have the armature and joint limits refer to an external set of data? For example, an XML file or a database?
There would also be the need to enable developers to build assessment tools that read the data for a manikin and provide feedback on potential risk. For example, this kind of tool may look at joint angles and the weight being lifted. Again, I understand this isn’t a simulation tool so the weight could be assigned to the manikin in the XML file.
It was a bit of challenge to understand where I should post my request. I’m trying to leverage the community members expertise to understand whether it would be possible to build a digital human in blender. My specific expertise is in ergonomics/human factors engineering.
Also, I’m not looking to profit from this specific effort. It’s just clear that there is a need for this kind of tool in the manufacturing/product development industries.
pkruithofjr:
I am an industrial engineering consultant. I had used asoftware package called Mannequin, now called HumanCad, to help clients"see" the limitations of human interaction in the work place.The first version of the software worked like a charm, but the company wassold. Each new version was more troublesome (and expensive) then the last and Ifinally gave up. The anthropomorphic data was generated from a US military database and categorizing the data by sex, body style, and nationality.
Should some enterprising individual have the time and talent,and use the suggestions suggested by the forum participants (Blender, MakeHuman,and the anthropomorphic data bases) and create pose-able figures importable into dfx and dwg formats, Iwould support his or her efforts and I am confident other engineering firma and architectural firms would as well.
Heck, I would even give them a spot on our company website!In my opinion it is a needed source of safety as well asa space visualization product.
I’m pretty curious and wanted to ask if something has happened in between?
In fact, I did an internship in a small company (usability industry) and they asked me to do something like that. I should use Blender and equip the human model with the given ergonomics values.
Unfortunately, a lot of small obstacles have arisen, where I now wonder whether it is even possible to create something like this in Blender.
So I would be really happy if I could get a little feedback. Also with regard to the programs you mentioned (Mannequin, HumanCad)