Tactile sensing is fundamental to human interaction, enabling the perception of pressure, texture, temperature, and pain through specialized sensory receptors embedded in the skin. Inspired by these biological mechanisms, tactile sensing has emerged as a key innovation in robotic applications, incorporating flexible and deformable sensors that can mimic human skin and the octopus’s mechanoreceptors. These bio-inspired sensors enhance tactile perception, allowing robots to detect interaction forces, surface properties and environmental conditions with high accuracy. The integration of tactile sensing technologies fosters the development of more intuitive, adaptive, and responsive robot systems. In addition to tactile sensing, proximity sensing plays a crucial role in the awareness of robot systems, which leads to human safety, enabling detection and avoidance of obstacles. Drawing inspiration from biological systems such as bat echolocation and electroreception in aquatic species, proximity sensors used in robots utilize ultrasonic waves, infrared detection, electromagnetic fields, and depth-sensing technologies. These sensing modalities improve the perception, planning, navigation, and control of robots more effectively in dynamic and unstructured environments. In this chapter, the integration of tactile and proximity sensing in robots is studied, particularly in the domain of collaborative and social robot applications, where precise sensing is essential for safe and effective human-robot interaction. In order to harness the potential of tactile and proximity sensors, embodied learning is used as a foundational framework for cognitive adaptation. By processing multimodal sensory inputs, robots can refine their perception, interaction strategies, and decision-making through experience. Additionally, machine learning techniques, particularly tensor-based algorithms, enhance this adaptability by enabling robots to interpret complex tactile and proximity sensor data efficiently. This chapter examines the role of embodied learning in leveraging tactile and proximity data for adaptive robot behaviour. Furthermore, case studies on collaborative, social robots and drones are presented to demonstrate how machine learning techniques optimize sensory processing for enhanced interaction, autonomy, and situational awareness.