G protein-coupled receptors (GPCRs) are the most abundant membrane proteins and the target of about 35% of approved drugs. Despite this, the structural basis of GPCR pharmacology is still a matter of intense study. Molecular dynamics (MD) simulations aim at expanding our knowledge of GPCR dynamics by building upon the recent advances in structural biology. However, the timescale limitations of classic MD hinder its applicability to numerous structural processes happening in time scales longer than microseconds (hidden structural transitions). For this reason, the overall MD impact on the study of GPCRs pharmacology and drug design is still limited. To overcome this, we have developed an unbiased adaptive sampling algorithm, namely multiple walker supervised MD (mwSuMD), and tested it on different hidden transitions involving GPCRs. By increasing the complexity of the simulated process, we report the binding and unbinding of the vasopressin peptide, the inactive-to-active transition of the glucagon-like peptide-1 receptor (GLP-1R), the stimulatory G protein (Gs) and inhibitory Gi binding to the adrenoreceptor β2 (β2 AR) and the adenosine 1 receptor (A1R) respectively, and the heterodimerization between the adenosine receptor A2 (A2AR) and the dopamine receptor D2 (D2R). We demonstrate that mwSuMD is a helpful tool for studying at the atomic level GPCR transitions that are challenging to address with classic MD simulations.