This paper presents an autonomous robotic platform for research of mechanically induced stress in plants growing in controlled greenhouse conditions. The platform provides a range of possibilities for mechanical stimuli including motion type, frequency, speed, and torque. The motions can be tailored for a single pot, making study of mechanical plant stress versatile, rapid and precise. We evaluate the performance of the platform for a use-case of basil plant cultivation. An eight week experiment was performed in greenhouse conditions on 220 basil plants. We show that the induction of mechanical stress by the platform significantly affects plant morphology, such as shortening stem length by 30 % -40 % and inter-node length by 50 % -80 %, while preserving leaf weight which is the main part of the basil plant used for culinary purposes. Results also show that variations in types of mechanical stimuli motions provides significant differences in the effect on plant morphology. Finally we show that decreasing the mechanical stimuli frequency to rates feasible to be performed manually significantly reduces the effect, stressing the need for autonomous systems capable of providing continuous stimuli during day and night. These results validate previously published findings in research of mechanical stress induction, and therefore implies the platform can be used for research of this phenomena.