[en] The volume of nuclear waste from operating nuclear reactors, decommissioning and military activity is ever-increasing worldwide. In 2018, the IAEA reported there existed some 6,317,000 m³ of legacy waste awaiting long-term storage. A method is required to categorize this waste ahead of this storage. Sort and Segregation could hold the answer to solving the problem of long-term waste disposal. Sort and Segregation enables the chemical and radiometric characterization of waste articles destined for long term nuclear waste storage, enabling nuclear waste management teams to subsequently store waste with confidence. Currently, Sort and Segregation is labor-intensive, delivers high worker dose rates and requires extensive user time input. An autonomous solution could, therefore, reduce the time and costs of the procedure, allowing for a high throughput. It could also increase the precision of the task, as it does not incur the excessive conservatisms that are introduced by human operators, as precautionary measures. Most importantly worker safety would be increased. A prototype autonomous sort and segregation system is demonstrated. It uses sensors and a robotic arm to facilitate the sorting and segregation of nuclear waste autonomously, using commercial off the shelf technology. This is an important proof of concept system which aims to demonstrate a solution to the problem of sorting and segregating nuclear waste for long term storage. A radiation sensor is used to characterize objects radiometrically. This collects full spectral data which can be used to identify the radioactive emitter present, alongside its radioactivity in dose rate and activity. Object mass is measured using a grasping arm. The robotic arm is fitted with a gripper which is used to grasp the objects. Given the classification of nuclear waste as a function of activity per unit mass, the grasping arm must weigh the object to make an estimation as to the correct waste stream in which the object should be placed. This is possible using the force torque sensors located in the robot's joints. The process for in-situ weighing on the robotic arm is explained and tested. (authors)