It's not accurate to say that no theory of integration on infinite-dimensional spaces exists. The Euclidean-signature Feynman measure has been constructed -- as a measure on a space of distributions -- in a number of non-trivial cases, mainly by the Constructive QFT school in the 70s.
The mathematical constructions reflect the physical ideas of effective quantum field theory: One obtains the measure on the space of field histories as the limit of a sequence/net of "regularized" integrals, which encode how the effective "long distance" degrees of freedom interact with each other after one averages out the short distance degrees of freedom in various ways. (You can imagine here that long/short distance refers to some wavelet basis, and that we get the sequence of regularized integrals by varying the way we divide the wavelet basis into short distance and long distance components.)
I don't think the main problem in the subject is that we need some new notion of integration. The Feynman measures we mathematicians can construct exhibit all the richness of the "higher categories" axioms, and moreover, the numerical computations in lattice gauge theory and in statistical physics indicates that the existing framework is at the least a very good approximation.
The problem, rather, is that we need a better way of constructing examples. At the moment, you have to guess which family of regularized integrals you ought to study when you try to construct any particular example. (In Glimm & Jaffe's book, for example, they simply replace the interaction Lagrangian with the corresponding "normally ordered" Lagrangian. In lattice gauge theory, they use short-distance continuum perturbation theory to figure out what the lattice action should be.)
Then -- and this is the really hard and physically interesting part -- you have to have enough analytic control on the family to say which observables (functions on the space of distributions) are integrable with respect to the limiting continuum measure. This is where you earn the million dollars, so to speak.
Most of this is standard theory of path integrals known to mathematical physicists so I will try to address all of your questions.
First let me say that the hypothesis you list for the action $S$ to make the path integral well defined, ie that $S=Q+V$ where $Q$ is quadratic and non-degenerate and $V$ is bounded are extremely restrictive. One should think of $V$ as defining the potential energy for interactions of the physical system and while it certainly true that one expects this to be bounded below, there are very few physical systems where this is also bounded above (this is also true for interesting mathematical applications...). Essentially requiring that the potential be bounded implies that the asymptotic behavior of $S$ in the configuration space is totally controlled by the quadratic piece. Since path integrals with quadratic actions are trivial to define and evaluate, it is not really that surprising or interesting that by bounding the potential one can make the integral well behaved.
Next you ask if anyone has studied the question of when an action $S$ gives rise to a well defined path integral: $ \int \mathcal{D}f \ e^{-S[f(x)]}$
The answer of course is yes. The people who come to mind first are Glimm and Jaffe who have made whole careers studying this issue. In all cases of interest $S$ is an integral $S=\int L$ where the integral is over your spacetime manifold $M$ (in the simplest case $\mathbb{R}^{n}$) and the problem is to constrain $L$. The problem remains unsolved but nevertheless there are some existence proofs. The basic example is a scalar field theory, ie we are trying to integrate over a space of maps $\phi: M \rightarrow \mathbb{R}$. We take an $L$ of the form:
$ L = -\phi\Delta \phi +P(\phi)$
Where in the above $\Delta$ is the Laplacian, and $P$ a polynomial. The main nontrivial result is then that if $M$ is three dimensional, and $P$ is bounded below with degree less than seven then the functional integral exists rigorously. Extending this analysis to the case where $M$ has dimension four is a major unsolved problem.
Moving on to your next point, you ask about another approach to path integrals called perturbation theory. The typical example here is when the action is of the form $S= Q+\lambda V$ where $Q$ is quadratic, $V$ is not, and $\lambda$ is a parameter. We attempt a series expansion in $\lambda$. The first thing to say here, and this is very important, is that in doing this expansion I am not attempting to define the functional integral by its series expansion, rather I am attempting to approximate it by a series. Let me give an example of the difference. Consider the following function $f(\lambda)$:
$f(\lambda)=\int_{-\infty}^{\infty}dx \ e^{-x^{2}-\lambda x^{4}}$
The function $f$ is manifestly non-analytic in $\lambda$ at $\lambda=0$. Indeed if $\lambda<0$ the integral diverges, while if $\lambda \geq 0$ the integral converges. Nevertheless we can still be rash and attempt to define a series expansion of $f$ in powers of $\lambda$ by expanding the exponential and then interchanging the order of summation and integration (illegal to be sure!). We arrive at a formal series:
$s(\lambda)=\sum_{n=0}^{\infty}\frac{\lambda^{n}}{n!}\int_{-\infty}^{\infty}dx \ e^{-x^{2}}(-x^{4})^{n}$
Of course this series diverges. However this expansion was not in vain. $s(\lambda)$ is a basic example of an asymptotic series. For small $\lambda$ truncating the series at finite order less than $\frac{1}{\lambda^{2}}$ gives an excellent approximation to the function $f(\lambda)$
Returning to the example of Feynman integrals, the first point is that the perturbation expansion in $\lambda$ is an asymptotic series not a Taylor series. Thus just as for $s(\lambda)$ it is misguided to ask if the series converges...we already know that it does not! A better question is to ask for which actions $S$ this approximation scheme of perturbation theory itself exists. On this issue there is a complete and rigorous answer worked out by mathematical physicists in the late 70s and 80s called renormalization theory. A good reference is the book by Collins "Renormalization." Connes and Kreimer have not added new results here; rather they have given modern proofs of these results using Hopf algebras etc.
Finally I will hopefully answer some of your questions about Chern-Simons theory. The basic point is that Chern-Simons theory is a topological field theory. This means that it suffers from none of the difficulties of usual path integrals. In particular all quantities we want to compute can be reduced to finite dimensional integrals which are of course well defined. Of course since we lack an independent definition of the Feynman integral over the space of connections, the argument demonstrating that it reduces to a finite dimensional integral is purely formal. However we can simply take the finite dimensional integrals as the definition of the theory. A good expository account of this work can be found in the recent paper of Beasley "Localization for Wilson Loops in Chern-Simons Theory."
Overall I would say that by far the currently most developed approach to studying path integrals rigorously is that of discretization. One approximates spacetime by a lattice of points and the path integral by a regular integral at each lattice site. The hard step is to prove that the limit as the lattice spacing $ a $ goes to zero, the so-called continuum limit, exists. This is a very hard analysis problem. Glimm and Jaffe succeeded in using this method to construct the examples I mentioned above, but their arguments appear limited. Schematically when we take the limit of zero lattice size we also need to take a limit of our action, in other words the action should be a function of $ a$. We now write $S(a)=Q+\lambda V+H(a,\lambda)$ Where as usual $Q$ is quadratic $V$ is not an $\lambda$ is a parameter. Our original action is $S=Q +\lambda V$
The question is then can we find an $H(a,\lambda)$ such that a suitable $a\rightarrow 0$ limit exists? A priori one could try any $H$ however the arguments of Glimm and Jaffe are limited to the case where $H$ is polynomial in $\lambda$. Physically this means that the theory is very insensitive to short distance effects, in other words one could modify the interactions slightly at short distances and one would find essentially the same long distance physics. It seems that new methods are needed to generalize to a larger class of continuum limits.
Best Answer
The path integral has many applications:
Mathematical Finance:
In mathematical finance one is faced with the problem of finding the price for an "option."
An option is a contract between a buyer and a seller that gives the buyer the right but not the obligation to buy or sell a specified asset, the underlying, on or before a specified future date, the option's expiration date, at a given price, the strike price. For example, an option may give the buyer the right but not the obligation to buy a stock at some future date at a price set when the contract is settled.
One method of finding the price of such an option involves path integrals. The price of the underlying asset varies with time between when the contract is settled and the expiration date. The set of all possible paths of the underlying in this time interval is the space over which the path integral is evaluated. The integral over all such paths is taken to determine the average pay off the seller will make to the buyer for the settled strike price. This average price is then discounted, adjusted for for interest, to arrive at the current value of the option.
Statistical Mechanics:
In statistical mechanics the path integral is used in more-or-less the same manner as it is used in quantum field theory. The main difference being a factor of $i$.
One has a given physical system at a given temperature $T$ with an internal energy $U(\phi)$ dependent upon the configuration $\phi$ of the system. The probability that the system is in a given configuration $\phi$ is proportional to
$e^{-U(\phi)/k_B T}$,
where $k_B$ is a constant called the Boltzmann constant. The path integral is then used to determine the average value of any quantity $A(\phi)$ of physical interest
$\left< A \right> := Z^{-1} \int D \phi A(\phi) e^{-U(\phi)/k_B T}$,
where the integral is taken over all configurations and $Z$, the partition function, is used to properly normalize the answer.
Physically Correct Rendering:
Rendering is a process of generating an image from a model through execution of a computer program.
The model contains various lights and surfaces. The properties of a given surface are described by a material. A material describes how light interacts with the surface. The surface may be mirrored, matte, diffuse or any other number of things. To determine the color of a given pixel in the produced image one must trace all possible paths form the lights of the model to the surface point in question. The path integral is used to implement this process through various techniques such as path tracing, photon mapping, and Metropolis light transport.
Topological Quantum Field Theory:
In topological quantum field theory the path integral is used in the exact same manner as it is used in quantum field theory.
Basically, anywhere one uses Monte Carlo methods one is using the path integral.