This paper provides a review of research on robotics in search and rescue operations in caverns, underground coal mines, disaster zones, and environments where the Global Navigation Satellite System (GNSS) is unavailable. The majority of applications for Simultaneous Localization and Mapping (SLAM), despite its maturity, are still restricted to indoor structured scenes or outside conventional weather conditions. The standard SLAM technology will experience malfunctions and perhaps fail as the robot encounters increasingly complicated environments. Three main issues that robots encounter in search and rescue (SAR) environments are outlined in this review, along with an introduction to the classic SLAM framework and a range of sensors used for SLAM. These issues include robots encountering increasingly complex terrain, changing environments and visibility, and autonomous exploration requirements. A thorough analysis of methods for resolving these problems is also included; they include multi-sensor fusion, active SLAM, and SLAM on uneven terrain. Also, a prospective path for future study is presented.