Despite the growing research interest in vulnerable road users, fine-grained cyclist datasets are still limited. Such datasets can provide empirical insights into previously unexplained cyclist behaviors as well as support calibration and validation of under-development cyclist behavior models. Current data collection methods primarily rely on aerial video footage, which is low-cost and easy to implement. However, it suffers from critical limitations such as data loss due to occlusion and limited spatial coverage. Sub-microscopic level details related to bicycle/riding dynamics and interactions with other road users are hard to capture. In response to these limitations, we develop an advanced privacy-preserving data collection platform from an ego bicycle perspective. Our instrumented bicycles are equipped with several ride dynamic sensors to precisely record cyclists' steering angle, pedaling power, cadence, and speed. Motion dynamic sensors, including a high-precision GPS module (with Real-Time Kinematic module) and an Inertial Measurement Unit, capture the cyclists' global positions, linear acceleration, angular velocity, and lean angle. Furthermore, we integrate LiDAR systems to detect surrounding infrastructures, and static and dynamic objects. A uniform Linux-based platform collects, synchronizes, and calibrates multi-frequency data streams from these diverse sensors. Precise motion dynamic data can describe complex dynamic models coupling lateral and longitudinal movements of cyclists instead of the point-mass model. LiDAR perception information provides interacting object characteristics such as object class, geometric profiles. Overall, Perception data serves as an environmental input that influences a cyclist's decision-making process. The dynamic riding data captured reflects the cyclist's responses, demonstrating the outcomes of their conscious or subconscious decisions. Precise motion dynamic data represent the execution of decisions and planning of cyclists. The comprehensive dataset provides insights into cyclist’s mental models, bridging perception, tactical decision-making, and operational execution within specific riding scenarios. Demonstration examples highlighting these capabilities will be showcased during the presentation.