VLN视觉语言导航算法研究员/专家
上海、深圳
全职
职位描述
1、负责VLN/VLM-Nav算法研发：视觉-语言-动作对齐、指令理解、路径规划与闭环执行； 2、攻关部署态关键问题：长距离指令、模糊指代（“那边/旁边/刚才看到的”）、动态人群、遮挡与重定位； 3、构建“语义导航”能力：语义地图/拓扑图、目标可达性判断、房间/区域识别、跨区域/跨楼层导航（如门禁/电梯）； 4、端到端闭环优化：导航成功率、平均耗时、卡死率、误入禁区率、舒适度（停走抖动/路径自然性）； 5、建立数据与评测体系：仿真与真机场景回放、对抗case库、自动化回归、线上失败归因与难例回流； 6、与SLAM、感知、多模态理解、任务规划、控制团队协作联调，形成可部署可迭代的导航能力栈；
职位要求
1、熟悉VLN/Embodied AI/Navigation相关研究与实现（任一：Transformer、policy learning、地图表征、跨模态对齐等）； 2、有项目/科研：导航、语义地图、目标可达性、指令理解等任一方向； 3、能在仿真环境跑通 baseline：训练、评测、可视化分析与失败归因。 加分项 - 具身导航数据合成/自动标注/仿真环境经验（Habitat/Gibson/Isaac等任一） - 真实机器人部署调优经验（人群、窄通道、玻璃反光、弱纹理等） - 顶会论文/专利/开源贡献（Embodied AI / Robotics / CV / ML相关）
投递
AGIBOT is an Embodied AI foundation model company developing both the intelligence layer and the corresponding robotic embodiments needed to bring general intelligence into the physical world. Founded in 2023, AGIBOT is building a full-stack Embodied AI platform across foundation models, data infrastructure, simulation, robotic hardware, and deployment. AGIBOT’s “Three Intelligences in One” architecture integrates Locomotion Intelligence, Interaction Intelligence, and Manipulation Intelligence into a unified embodied system. AGIBOT’s portfolio spans humanoid robots, quadrupeds, dexterous systems, and commercial cleaning solutions. In March 2026, AGIBOT announced that its 10,000th robot had rolled off the production line.
