This thesis addresses the challenge of autonomous navigation in partially known environments, focusing on mobile robots equipped with limited sensing capabilities. Path planning and sensor usage are closely linked in such environments, as each sensor activation incurs a cost. Thus, the robot faces a trade-off: it can either follow longer, sensor-free paths or activate sensors to take shorter, more efficient routes. To address this, we develop two joint decision-making frameworks integrating path planning with strategic sensor activations to optimize navigation efficiency under resource constraints. A Regret-aware Joint Sensing and Path Planning presents a combined sensing and control framework designed to minimize path length while efficiently using sensing. This methodology employs a joint cost map to evaluate the value of the information gained from each potential sensor activation, prioritizing sensing only where substantial information gain or path improvement is likely. Initial results demonstrate that this approach allows the robot to allocate sensors effectively, avoiding unnecessary activations and improving navigation efficiency compared to simpler controllers. A Sensor-aware Planner and Regret-based Cost Function method builds on the first by expanding the robot’s planning capability to incorporate sensor use as part of the path planner’s state space. This methodology includes a modified A* algorithm that operates within an enlarged, multi-dimensional search space, representing the environment and the robot’s real-time available sensing budget. An edge-based cost function dynamically evaluates local decisions based on regret and information value. The proposed sensor-aware A* planner enables the robot to anticipate future sensor activations, strategically utilizing the sensor across its path. Repeated simulation iv shows statistical evidence that this approach enhances the robot’s ability to prioritize critical sensing locations, improving path efficiency while reducing overall sensor usage. Together, these methodologies advance the robot’s capacity to navigate unknown environments by balancing short-term and long-term sensing strategies, providing a scalable framework for resource-limited exploration in complex, partially mapped environments.