Large language models have found utility in the domain of robot task planning and task decomposition. Nevertheless, the direct application of these models for instructing robots in task execution is not without its challenges. Limitations arise in handling more intricate tasks, encountering difficulties in effective interaction with the environment, and facing constraints in the practical executability of machine control instructions directly generated by such models. In response to these challenges, this research advocates for the implementation of a multi-layer large language model to augment a robot’s proficiency in handling complex tasks. The proposed model facilitates a meticulous layer-by-layer decomposition of tasks through the integration of multiple large language models, with the overarching goal of enhancing the accuracy of task planning. Within the task decomposition process, a visual language model is introduced as a sensor for environment perception. The outcomes of this perception process are subsequently assimilated into the large language model, thereby amalgamating the task objectives with environmental information. This integration, in turn, results in the generation of robot motion planning tailored to the specific characteristics of the current environment. Furthermore, to enhance the executability of task planning outputs from the large language model, a semantic alignment method is introduced. This method aligns task planning descriptions with the functional requirements of robot motion, thereby refining the overall compatibility and coherence of the generated instructions. To validate the efficacy of the proposed approach, an experimental platform is established utilizing an intelligent unmanned vehicle. This platform serves as a means to empirically verify the proficiency of the multi-layer large language model in addressing the intricate challenges associated with both robot task planning and execution.

Spatial intelligence requires agents to form and utilize internal representations of the physical world for perception, reasoning, and generation. While recent advances in foundation models, embodied systems, and three-dimensional representation learning have substantially expanded spatial capabilities, existing research remains fragmented across heterogeneous tasks and model paradigms. This survey revisits spatial intelligence from a cognitive map perspective and positions cognitive maps as its representational blueprint. In this view, diverse lines of research can be understood through a shared question: how an internal spatial representation is constructed, maintained, reasoned over, and realized. To make this perspective operational, we define cognitive maps as internal spatial representations characterized by abstraction, globality, and persistency. Based on this definition, we organize the literature into three cognitive-map-centric processes that correspond to the core dimensions of spatial intelligence: perception for cognitive map construction, reasoning for internal inference with the map, and generation for external realization of the map. By adopting a mechanism-centric viewpoint, this survey connects previously isolated research directions into a coherent framework and identifies emerging challenges toward unified spatial intelligence systems. The related resources of this study are accessible at https://github.com/Klingsor-tyx/Awesome-Spatial-Cognitive-Map.

Source-free domain generalization (SFDG) tackles the challenge of adapting models to unseen target domains without access to source domain data. To deal with this challenging task, recent advances in SFDG have primarily focused on leveraging the text modality of vision-language models such as CLIP. These methods involve developing a transferable linear classifier based on diverse style features extracted from the text and learned prompts or deriving domain-unified text representations from domain banks. However, both style features and domain banks have limitations in capturing comprehensive domain knowledge. In this work, we propose Prompt-Driven Text Adapter (PromptTA) method, which is designed to better capture the distribution of style features and employ resampling to ensure thorough coverage of domain knowledge. To further leverage this rich domain information, we introduce a text adapter that learns from these style features for efficient domain information storage. Extensive experiments conducted on four benchmark datasets demonstrate that PromptTA achieves state-of-the-art performance. The code is available at https://github.com/zhanghr2001/PromptTA.