Total knee arthroplasty (TKA) is an effective surgical intervention for end stage knee osteoarthritis in elderly patients, with emerging robotically assisted techniques aiming to enhance surgical precision and patient outcomes. This study aimed to compare medium-term balance and functional outcomes between robotically assisted and conventional manual TKA in community-dwelling elderly patients. This cross-sectional study included 50 elderly patients undergoing TKA, who were divided into robotically assisted (n = 25) and conventional manual (n = 25) groups. Demographic and clinical data, balance performance, and functional outcomes were compared at nearly 1.5 years postoperatively. Outcome measures included balance performance assessed by the Berg Balance Scale (BBS), pain via the Visual Analog Scale (VAS), knee function as measured by the Lysholm Knee Scoring Scale, quality of life using the Short Form-12 (SF-12), joint awareness as evaluated by the Forgotten Joint Score-12 (FJS-12), and surgical satisfaction. The groups had similar demographic and clinical data regarding age, gender, follow-up duration, surgical time, and anesthesia type ( > 0.05). The robotically assisted group demonstrated better balance performance on the BBS ( = 0.043) and had a statistically shorter length of hospital stay (1.22 vs. 1.42 days; = 0.005). However, no statistically significant differences were observed in VAS activity pain ( = 0.053), Lysholm Knee Scoring Scale ( = 0.117), SF-12 physical and mental scores ( = 0.174 and = 0.879), FJS-12 ( = 0.760), and surgical satisfaction ( = 0.218). Robotically assisted TKA is associated with advantageous postoperative recovery, particularly in terms of balance performance, showing no clinical difference in other functional outcomes compared to the conventional manual technique. From a physical therapy perspective, these findings emphasize the importance of developing tailored and effective rehabilitation strategies in the medium term for functional recovery in the elderly population.