Intra-row weeding is a time-consuming and technically demanding operation in orchard and vineyard management, requiring high precision to avoid damage to crop trunks while ensuring effective weed removal. To address this challenge, a rotary weeder implement integrated with an autonomous electric robot was developed and evaluated. The conventional hydraulic drive system was replaced with an electric motor, and selected mechanical components were redesigned to reduce the overall weight of the implement. Side-shift, height, and tilt adjustments were achieved using linear electric actuators to enable precise positioning during operation. For trunk detection, two sensing approaches were investigated, namely a conventional electromechanical feeler sensor and a sonar-based sensor. Autonomous row following was accomplished using data acquired from a two-dimensional laser scanner. The robotic prototype was experimentally evaluated at a forward speed of 0.16 m s⁻¹ and a working depth of 40 mm. The overall performance of both trunk detection systems was assessed in terms of weeding quality and power consumption. The experimental results demonstrated that the autonomous intra-row weeding robot could serve as a viable alternative to conventional machinery. Among the tested sensing systems, the sonar sensor exhibited superior performance compared to the adjusted feeler mechanism. The integration of autonomous navigation with electric weeding technology showed potential to improve weeding quality while reducing power requirements in future field applications.