Background Clinical practice guidelines recommend that exercise is essential in the self-management of ankylosing spondylitis (AS). Attending supervised interventions requiring periodic medical center visits can be difficult and patients may decline participation, whereas effective home-based exercise interventions that do not require regular medical center visits are likely to be more accessible for AS patients. Objective The goal of the research was to investigate the adherence, efficacy, and safety of a wearable technology–assisted combined home-based exercise program in AS. Methods This was a 16-week investigator-initiated, assessor-blinded, randomized, pilot controlled trial conducted at Chinese People’s Liberation Army General Hospital. We enrolled patients with AS who had no regular exercise habits and had been stable in drug treatment for the preceding month. Patients were randomly assigned (1:1) using a computer algorithm. An exercise program consisting of moderate-intensity aerobic exercise and functional exercise was given to the patients in the intervention group. The exercise intensity was controlled by a Mio FUSE Heart Rate Monitor wristband, which uses photoplethysmography to measure heart rate. Patients in the control group received usual care. The primary outcome was the difference in the Ankylosing Spondylitis Disease Activity Score (ASDAS). The secondary outcomes were patient global assessment (PGA), physician global assessment (PhGA), total pain, nocturnal pain, Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), BAS Functional Index (BASFI), BAS Metrology Index (BASMI), Spondyloarthritis International Society Health Index (ASAS HI), 36-item Short Form Survey (SF-36), maximal oxygen uptake (VO2) max, body composition, range of motion of joints, and muscle endurance tests. Retention rate, adherence rate, barriers to being active, and adverse events were also assessed. Results A total of 77 patients were screened, of whom 55 (71%) patients were enrolled; 2% (1/55) withdrew without treatment after randomization. Patients were assigned to the intervention (n=26) or control group (n=28). The median adherence rate of the prescribed exercise protocol was 84.2% (IQR 48.7%-97.9%). For the primary outcome, between-group difference of ASDAS was significant, favoring the intervention (–0.2, 95% CI –0.4 to 0.02, P=.03). For the secondary outcomes, significant between-group differences at 16 weeks were detected in PGA, PhGA, total pain, BASDAI, BASDAI-fatigue, BASDAI–spinal pain, BASDAI–morning stiffness intensity, BASFI, and BASMI. Moreover, the frequency of difficulty in ASAS HI-motivation at 16 weeks was less in the intervention group (P=.03). Between-group difference for change from baseline were also detected in VO2 max, SF-36, back extensor endurance test, and the range of motion of cervical lateral flexion at 16 weeks. Lack of time, energy, and willpower were the most distinct barriers to being active. Incidences of adverse events were similar between groups (P=.11). Conclusions Our pilot study suggests that this technology-assisted combined home-based exercise program can improve the clinical outcomes of patients with AS who have no exercise habit, with good adherence and safety profile. Trial Registration Chinese Clinical Trial Registry ChiCTR1900024244; http://www.chictr.org.cn/showproj.aspx?proj=40176