I highlighted the some of your attempts:
$\ce{H_d}$ has $\mathrm pK_\mathrm{a} \approx 9$, so it will not be the most acidic.
This assumption is correct. For aspartic acid, $\mathrm pK_\mathrm{a}(\ce{H3N+}) = 9.6$.
$\ce{NH3+}$ has very high -I effect but no mesomeric effect, so $\ce{H_c}$ appears to be quite acidic. But at the same time it is at the alpha position of the carboxylic acid and may not be effectively conjugated in the $\ce{C=O}$ bond.
This assumptions are correct as well. The $\alpha$-hydrogens in acids and esters are not as acidic as those in aldehydes and ketones because carbonyl groups of them already involved woth attached $\ce{OH}$ or $\ce{OR}$ groups, respectively. However, $\alpha$-hydrogen in amino acids fairly acidic than those in aldehydes or ketones due to strong -I effect from $\ce{H3N+\!-}$ group.
Here can the electron pair if formed by loss of $\ce{H_c}$ be conjugated on the $\ce{C=O}$ bond?
Yes, that is the main reason it has being acidic (compared to $\beta$-hydrogen or other $\ce{C-H}$ in the same alkyl chain). However, it won't leave to make a carboanion until it has reached the correct $\mathrm{pH}$. The $\mathrm pK_\mathrm{a}$ of $\alpha$-hydrogen in an $\alpha$-amino acid is in the range of $16$ to $17$ (Ref.1). The exception is been D-phenylglycine, $\ce{Ph-CH(NH2)-CO2H}$ $(\mathrm pK_\mathrm{a} = 14.9)$, which is not a natural essential amino acid. The reason for lower $(\mathrm pK_\mathrm{a}$ value is the additional resonance contribution from neighboring $\ce{Ph}$ group. The few examples are given below:
$$
\alpha\text{-Hydrogen $\mathrm pK_\mathrm{a}$ values (Ref.1)}\\
\begin{array}{c|ccc}
\hline
\text{Amino acid} & \mathrm pK_\mathrm{a} \\
\hline
\text{Alanine} & 16.5 \\
\text{Valine} & 17.0 \\
\text{Isoleucine} & 16.9 \\
\text{Leucine} & 16.7 \\
\text{Phenylalanine} & 16.2 \\
\text{D-Phynylglicine} & 14.9 \\
\hline
\end{array}
$$
Therefore, it is safe to assume that $\alpha$-hydrogen of acid is the least acidic among other acidic hydrogen in the molecule, even lower than that of $\ce{H3N+\!-}$ group.
$\ce{H_b}$ may be acidic but no special effect may be attributed to it, similar to $\ce{H_a}$.
This assumption is incorrect. As a matter of fact, $\ce{H_a}$ and $\ce{H_b}$ are the most acidic hydrogens among other acidic hydrogens in the molecule (this was already proved in the other answer, so that I'm not going to repeat it again). Between the two, $\ce{H_b}$ is more acidic than $\ce{H_a}$ since it is closer to $\ce{H3N+\!-}$ group and feel more -I effect (through less numbers of $\sigma$-bonds). The actual $\mathrm pK_\mathrm{a}$ of $\ce{H_a}$ and $\ce{H_b}$ are $3.65$ and $1.88$, respectively.
Thus, the correct order should be $\ce{H_b} \gt \ce{H_a} \gt \ce{H_d} \gt \ce{H_c}$.
Reference:
- Eric D. Stroud, Dennis J. Fife, Grant Gill Smith, "A method for the determination of the $\mathrm pK_\mathrm{a}$ of the $\alpha$-hydrogen in amino acids using racemization and exchange studies," J. Org. Chem. 1983, 48(26), 5368–5369 (https://doi.org/10.1021/jo00174a045).