Reactive control is often considered insufficient for multi-objective tasks because conflicting objectives give rise to local minima. We argue this limitation is not inherent but arises from static encodings that fail to reflect how objectives currently interact. We exploit the interaction structure encoded in a graph-based world model by extending it with nullspace projections: conflicts are resolved where they arise by projecting lower-priority gradients into the nullspace of higher-priority ones, with priorities determined continuously from the current state. We demonstrate this in two domains where conflicts between objectives are central: navigation around non-convex obstacles, where static potential fields fundamentally fail, and planar pushing of non-convex objects, where our method achieves $100\%$ success across one-hundred configurations versus $0\%$ for the steepest-descent baseline and ${\sim}55\%$ for diffusion policy, without demonstrations or retraining. The same formulation transfers directly to a real robot with additional perceptual and kinematic constraints, accommodating them through the same mechanism.

翻译：反应控制通常被认为不足以应对多目标任务，因为冲突目标会导致局部极小值。我们认为这种局限性并非固有，而是源于无法反映目标当前相互作用的静态编码。我们利用基于图的世界模型中编码的交互结构，通过扩展零空间投影来破解这一局限：将低优先级梯度投影到高优先级梯度的零空间中，在冲突产生之处予以解决，而优先级则根据当前状态持续确定。我们在两个以目标冲突为核心问题的领域进行了验证：在非凸障碍物周围的导航任务中，静态势场从根本上失效；在非凸物体的平面推动任务中，我们的方法在一百种配置下实现了100%的成功率，而最陡下降基线方法为0%，扩散策略约为55%，且无需演示或重新训练。相同的方案可直接迁移至具有额外感知与运动学约束的真实机器人，通过同一机制即可适配这些约束。