CALM: Current Aligned Link Manipulation for Single Arm Oversized Object Lifting

Hu, Jun, Chen, Sihan, Jovanovic, Kosta, Navarro-Alarcon, David, Wang, Xueqian, Pan, Jia, Zhou, Peng · N/A · 2026

Summary

Most robots manipulate objects solely with their end effectors, whereas humans flexibly leverage different body parts, such as the forearm and elbow, especially when handling oversized objects.

Abstract Summary

Most robots manipulate objects solely with their end effectors, whereas humans flexibly leverage different body parts, such as the forearm and elbow, especially when handling oversized objects. Learning such whole-arm manipulation is chal-lenging due to long-horizon sparse rewards, limited contact sens-ing, and the sim-to-real gap in contact and actuator dynamics. To address these challenges, we propose Current-Aligned Link Manipulation, a framework for learning long-horizon contact-rich manipulation using motor current as joint load related feedback. Three stage-specific policies first learn repositioning, grasping, and lifting using privileged simulation information, and a stage router sequences them to generate complete task demonstrations. For sim-to-real transfer, a causal current mapper predicts physical motor current from simulated joint histories, aligning the actuator current observation between simulation and hardware. A unified student policy then learns from these demonstrations using only deployable sensor observations and is further refined with DAgger. The task policies are trained entirely in simulation, and the final student is deployed on hardware. Experiments demonstrate 76.2% (762/1000 trials) complete-task success in simulation and 73.3% success (22/30 trials) on the physical robot for sequential oversized-object lifting.

Key Points

Most robots manipulate objects solely with their end effectors, whereas humans flexibly leverage different body parts, such as the forearm and elbow, especially when handling oversized objects.
Learning such whole-arm manipulation is chal-lenging due to long-horizon sparse rewards, limited contact sens-ing, and the sim-to-real gap in contact and actuator dynamics.
To address these challenges, we propose Current-Aligned Link Manipulation, a framework for learning long-horizon contact-rich manipulation using motor current as joint load related feedback.
Three stage-specific policies first learn repositioning, grasping, and lifting using privileged simulation information, and a stage router sequences them to generate complete task demonstrations.
For sim-to-real transfer, a causal current mapper predicts physical motor current from simulated joint histories, aligning the actuator current observation between simulation and hardware.

Method

We propose CALM to address the task of lifting oversized boxes with a single robot arm. The overall framework is illustrated in Fig. 2 .

The framework comprises six components: (1) Causal Current Alignment maps simulated histories to predicted physical current observations; (2) Privileged Teacher Training trains three teachers using privileged simulation state; (3) Stage Conditioned Rollout Collection routes the teachers to generate complete trajectories; (4) Student Distillation trains one policy to imitate the routed teacher behaviors using o...

Method — CALM: Current Aligned Link Manipulation for Single Arm Oversized Object Lifting — Fig. 3: Experimental setups. (a) Isaac Lab simulation environment. (b) Physical robot setup, where two cameras detect ArUco markers attached to the box to track its position.

Experiments and Results

The experiments are designed to answer three research questions: (1) Does current alignment reduce the observation gap between simulation and reality? (2) Do stage aware distillation and DAgger improve unified student performance? (3) How reliably does the complete CALM system perform physical box lifting? These questions and the corresponding experiments are summarized in Table III .

The simulation and physical experiments use identical six axis Universal Robots UR7e robots.

Experiments and Results — CALM: Current Aligned Link Manipulation for Single Arm Oversized Object Lifting — Fig. 4: Offline current alignment. (a) Raw simulated joint effort (right axes), recorded physical motor current, and current predictions (left axes) over the first eligible continuous 15 s interval. (b) MAE for each joint and the mean across joints over all 19,213 held out samples. The simulated joint effort baseline is converted to amperes using a separate linear scaling for each joint fitted on the training data.

Sources and Demonstrations

Figures are reproduced from Hu et al. See the full paper for experimental details and the project page, when the paper links one, for demonstrations and videos.