cs.ROFeb 9, 2026
SaveLearning to Drive in New Cities Without Human Demonstrations
Organizations: WhiRL, University of Oxford · FLAIR, University of Oxford · Delft University of Technology · NYU Tandon School of Engineering
Abstract
While autonomous vehicles have achieved reliable performance within specific operating regions, their deployment to new cities remains costly and slow. A key bottleneck is the need to collect many human demonstration trajectories when adapting driving policies to new cities that differ from those seen in training in terms of road geometry, traffic rules, and interaction patterns. In this paper, we show that self-play multi-agent reinforcement learning can adapt a driving policy to a substantially different target city using only the map and meta-information, without requiring any human demonstrations from that city. We introduce NO data Map-based self-play for Autonomous Driving (NOMAD), which enables policy adaptation in a simulator constructed based on the target-city map. Using a simple reward function, NOMAD substantially improves both task success rate and trajectory realism in target cities, demonstrating an effective and scalable alternative to data-intensive city-transfer methods. Project page: https://zerlinwang.github.io/NOMAD/
Figures & tables
Figure 1: Zero-shot transfer versus NOMAD.
Figure 2: NOMAD overview . Starting from a source-city imitation policy , NOMAD adapts it to a target city using map segments and meta-information . A scenario generator samples initial states and goals. The generated scenarios are loaded in a data-driven multi-agent driving simulator, yielding a simulator of . The policy is initialized from and optimized via KL-regularized self-play MARL. Checkpoints along training yield an adapted policy set , from which practitioners select a deployment policy to suit their preferences.
Figure 3: NOMAD consistently improves the success–realism Pareto frontier in unseen target cities. We plot mean success rate versus mean realism meta score over 5 runs for three transfer settings: (a) Boston-to-Singapore (primary), (b) Singapore-to-Boston, and (c) Singapore-to-Pittsburgh. Each dot is a NOMAD training checkpoint, colored by the cumulative number of interaction steps; the red dashed curve denotes the empirical Pareto frontier over NOMAD checkpoints. Stars denote BC reference policies, while dashed curves denote RL-based ablations initialized or trained under different data/settings.
| Policy | Realism | Success |
|---|---|---|
| 0.8056 | 88.26% | |
| Random | 0.4074 | 4.30% |
| Constant Velocity | 0.6147 | 19.45% |
| 0.6795 | 42.25% | |
| BC (Singapore) | 0.7011 | 55.49% |
| NOMAD Frontier | 0.7570 0.7697 | 87.28 94.27% |
Table 1: NOMAD substantially improves both success and realism over zero-shot transfer and imitation baselines. We report success rate and realism meta score for representative policies in closed-loop Singapore evaluation. denotes a policy obtained by inferring actions from logged expert trajectories and discretizing them to match the action space; its realism meta score serves as an approximate upper bound under the current POSG formulation.
Figure 4: Behavioral priors are critical for preserving kinematically realistic driving during self-play. We compare the kinematic realism score of NOMAD against self-play from scratch in the Boston-to-Singapore transfer setting. Self-play without behavioral priors struggles to learn kinematically realistic behavior, while NOMAD preserves substantially more realistic motion patterns despite lacking explicit kinematic rewards.
| Transfer | Policy | Failure | Off-road only | Collision only | Both | Neither (timeout) |
|---|---|---|---|---|---|---|
| Boston Singapore | 57.7% | 33.3% | 1.2% | 2.7% | 20.5% | |
| NOMAD | 7.7% 1.8 | 0.6% 0.2 | 0.4% 0.2 | 0.3% 0.3 | 6.5% 1.7 | |
| Singapore Boston | 63.3% | 18.8% | 3.1% | 8.1% | 33.3% | |
| NOMAD | 13.6% 5.6 | 1.3% 0.3 | 1.5% 0.5 | 1.1% 0.3 | 9.7% 4.3 |
Table 2: NOMAD eliminates most failures in the target city. We report the failure rate (%) and its breakdown by failure mode in closed-loop evaluation for Boston-to-Singapore and Singapore-to-Boston transfer. NOMAD results are reported as mean standard deviation over 5 runs.
Figure 5: Qualitative comparison of the zero-shot policy and NOMAD under Singapore-to-Boston transfer. Both rows start from the same scenario and initial state. Numbers identify the controlled vehicles and marks the goal of vehicle ; a controlled vehicle is drawn in red while it is in collision and in orange while it is off-road. More examples are in Appendix O .
Appendix figures & tables49 assets
Supplementary material from the paper’s appendix.
Appendix
Figure 6: Policy Network architecture. The policy consists of modality-specific embedding layers for ego state, surrounding agents, and road observations, followed by a shared observation encoder. The encoded features are passed to separate actor and critic branches with residual blocks. The actor outputs trajectory increments , while the critic predicts the scalar state value .
Figure 7: Visualization of the ego-centric observation space.
| Hyperparameter | Value |
|---|---|
| Observation Dim | 2984 |
| Observation Embedding Dim | 64 3 |
| Encoder Width | 512 |
| Dropout | 0.01 |
| Residual Block Num | 1 |
| Projection Width | 256 |
Table 3: Network hyperparameters.
| Hyperparameter | Value |
|---|---|
| Observation Radius | 100.0 |
| Max Controllable Agents Num | 64 |
| Max Road Observation Num | 200 |
| Ego observation | 1 6 |
| Agents Observation | (64-1) 6 |
| Road Observation | 200 13 |
Table 4: Environment configurations.
| Hyperparameters | Value |
|---|---|
| Dataset Scenario Num | 3200 |
| Batch Size | 64 |
| Loss Function | Cross Entropy |
| Learning Rate | 1e-3 |
| Weight Decay | 1e-2 |
| Scheduler | Cosine Annealing |
Table 5: Behavior Cloning hyperparameters.
| Hyperparameters | Value |
|---|---|
| World Num | 1,000 |
| Batch Size | 8,192 |
| Buffer Size | 131,072 |
| Update Epoch | 4 |
| Discount Factor | 0.99 |
| GAE Discount Factor | 0.95 |
Table 6: PPO hyperparameters.
| City | |||
|---|---|---|---|
| Boston | 1.4694 | 0.1897 | 0.4252 |
| Pittsburgh | 2.1327 | 0.0000 | 0.3982 |
| Singapore | 1.2630 | 0.1363 | 0.2500 |
Table 7: City-specific traffic density parameters for (vehicles/km), where is the shape and is the scale parameter.
| Category | Parameter | Value |
|---|---|---|
| Vehicle Placement | Minimum time gap | |
| Minimum distance | ||
| Static vehicle probability | ||
| Speed Distribution | Mean speed fraction | |
| Speed std fraction | ||
| Goal Generation | Distance mean coefficient |
Table 8: Shared scenario generation parameters.
| Generated | Logged | Held-out eval | |
| (training) | (reference) | (evaluation) | |
| # scenes | 25,600 | 3,200 | 800 |
| Road geometry | |||
| Lane length (km) | 9.9 [7.0, 13.0] | 9.9 [7.0, 13.0] | 10.1 [7.5, 13.0] |
| # lane segments | 401 [283, 490] | 401 [283, 490] | 410 [312, 495] |
| # crosswalks | 19 [13, 26] | 19 [13, 26] | 20 [14, 26] |
Table 9: Scenario statistics for Boston Singapore. All statistics are computed on the initial scene ( ). Each entry is the median over scenes, followed in brackets by the interquartile range (IQR), i.e., the 25th and 75th percentiles, which span the middle 50% of scenes. The row marked † reports the mean instead, as its median is 0 in all sets. Generated scenes reuse the training-split maps. The classifier AUC is computed on the 9 traffic features against the held-out evaluation set and reported as mean standard deviation over 5 random draws.
Figure 8: Pareto frontier of success rate versus realism meta score across training checkpoints for Boston-to-Singapore transfer with small traffic density perturbation. Each point corresponds to a policy checkpoint, colored by the number of interaction steps. The yellow star denotes the zero-shot transfer policy from Boston, while the lime star shows behavior cloning using Singapore trajectories. The red dashed line and points indicate the empirical Pareto frontier on this test set.
Figure 9: Pareto frontier of success rate versus realism meta score across training checkpoints for Boston-to-Singapore transfer with large traffic density perturbation. Each point corresponds to a policy checkpoint, colored by the number of interaction steps. The yellow star denotes the zero-shot transfer policy from Boston, while the lime star shows behavior cloning using Singapore trajectories. The red dashed line and points indicate the empirical Pareto frontier on this test set.
Figure 10: Visualization of scenarios from different cities in the nuPlan dataset. The rectangles denote vehicles and the circles with the corresponding color denote their goals. To mirror the Singapore and make it a right-hand traffic city, we first flip the scenario across x-axis, then we change the orientation of road edges while keeping the orientation of road lanes.
Figure 11: Examples of generated scenarios on Singapore map for two different sampled map segments.
| Policy | Scenarios | Realism Meta Score | Success Rate |
|---|---|---|---|
| Boston | 0.8311 | 100.0% | |
| Singapore | 0.8581 | 100.0% | |
| Boston | 0.7814 | 85.47% | |
| Singapore | 0.8056 | 88.26% |
Table 10: Performance of expert policies in different cities.
Figure 12: Detailed sub-metrics in realism metric and ADE in the experiment of transferring from Boston to Singapore. Although the BC policy (at the beginning of training) achieves reasonable kinematic smoothness, it frequently violates map constraints (e.g., off-road driving), leading to a poor map-based realism meta score. In our experiments, behavior cloning serves as a canonical source-city planner. While its absolute performance depends on dataset coverage, our focus is on the performance degradation induced by the lack of supervised finetuning in the target city and the ability of NOMAD to mitigate it without demonstrations.
Figure 13: Detailed sub-metrics in realism metric and ADE in the experiment of transferring from Singapore to Boston.
Figure 14: Detailed sub-metrics in realism metric and ADE in the experiment of transferring from Singapore to Pittsburgh.
Figure 15: Detailed realism sub-metrics and ADE for behavior cloning followed by self-play MARL in Boston, evaluated zero-shot in Singapore.
Figure 16: Comparison of all metrics between self-play with and without behavioral priors. Behavioral priors play a critical role in stabilizing self-play training and guiding exploration toward human-like driving under the supervision of a simple reward function.
Figure 17: Source-city self-play is beaten by NOMAD. The dashed red curve (BC (Boston) + RL (Boston)) denotes policies trained with behavior cloning and self-play in Boston only, evaluated zero-shot in Singapore. The dashed purple curve denotes NOMAD. Source-city self-play improves over zero-shot transfer but fails to approach the Pareto frontier achieved by NOMAD, demonstrating that self-play must be grounded in the target-city map to effectively adapt across cities.
Figure 18: NOMAD closes most of the gap between zero-shot transfer and target-city demonstration-accessed policy. We compare zero-shot transfer, NOMAD (RL with generated scenarios), and a policy trained with 3,200 scenarios with logged trajectories in the target-city using the same BC+RL training protocol (BC (Singapore) + RL (Singapore)). While the policy with access to target-city demonstrations yields additional gains over NOMAD, this improvement is modest compared to the substantial gap between zero-shot transfer and NOMAD, highlighting the effectiveness of map-based self-play for city transfer.
Figure 19: Realism meta score and success rate over interaction steps on the Singapore test scenarios. Results are shown with mean and 95% confidence interval over 5 seeds.
Figure 20: Realism meta score and success rate during map-based self-play without behavioral priors in Singapore. Self-play without pretraining and regularization achieves near-perfect success but a lower realism meta score. Shaded regions indicate 95% confidence interval across 5 runs.
Figure 21: The KL regularization weight controls the success–realism trade-off during adaptation. We compare Pareto frontiers with different . Smaller KL weights favor higher success at the cost of realism, while larger KL weights encourage more realistic behaviors but constrain success rate.
Figure 22: Pareto frontier of success rate versus realism meta score across training checkpoints for Singapore with log-replay sim agents. Each point corresponds to a policy checkpoint, colored by the number of interaction steps. The yellow star denotes the zero-shot transfer policy from Boston, while the lime star shows behavior cloning using Singapore trajectories. The red dashed line and points indicate the empirical Pareto frontier on this test set. Frontier checkpoints consistently dominate the baselines and exhibit only a mild trade-off between task success and realism.
Figure 23: Pareto frontier of success rate versus realism meta score across training checkpoints for Boston with log-replay sim agents. Each point corresponds to a policy checkpoint, colored by the number of interaction steps. The yellow star denotes the zero-shot transfer policy from Singapore, while the lime star shows behavior cloning using Boston trajectories. The red dashed line and points indicate the empirical Pareto frontier on this test set. Frontier checkpoints consistently dominate the baselines and exhibit only a mild trade-off between task success and realism.
Figure 24: Pareto frontier of success rate versus realism meta score across training checkpoints for Pittsburgh with log-replay sim agents. Each point corresponds to a policy checkpoint, colored by the number of interaction steps. The yellow star denotes the zero-shot transfer policy from Singapore, while the lime star shows behavior cloning using Pittsburgh trajectories. The red dashed line and points indicate the empirical Pareto frontier on this test set. Frontier checkpoints consistently dominate the baselines and exhibit only a mild trade-off between task success and realism.
| Checkpoint | Realism Self-Play | Realism Cross-Play | Success Rate Self-Play | Success Rate Cross-Play |
|---|---|---|---|---|
| 3690 | 0.768 0.004 | 0.769 0.003 | 0.885 0.011 | 0.885 0.008 |
| 3710 | 0.771 0.003 | 0.771 0.003 | 0.873 0.011 | 0.872 0.010 |
| 3740 | 0.770 0.003 | 0.769 0.002 | 0.879 0.017 | 0.880 0.011 |
| 3930 | 0.768 0.003 | 0.767 0.002 | 0.897 0.015 | 0.894 0.013 |
| 3980 | 0.766 0.004 | 0.766 0.003 | 0.900 0.014 | 0.902 0.010 |
| 4010 | 0.766 0.005 | 0.766 0.003 | 0.901 0.011 | 0.899 0.009 |
Table 11: Self-play vs cross-play summary at Pareto-frontier checkpoints (Boston Singapore). We evaluate each selected frontier checkpoint under both self-play (paired with the same-seed partner) and cross-play (paired with different-seed partners) using independently trained seeds. We report mean standard deviation across pairings. Cross-play success and realism closely match self-play at the same checkpoint, indicating that NOMAD does not rely on seed-specific conventions.
Figure 25: Cross-play success matrices across Pareto-frontier checkpoints (Boston Singapore). Each panel corresponds to one selected checkpoint on the success–realism Pareto frontier. Within each panel, entry reports the ego success rate when pairing seed with seed ( total seeds). Diagonal entries are self-play (same seed paired with itself), while off-diagonals are cross-play (different seeds). Off-diagonal performance is comparable to diagonal performance across checkpoints, indicating minimal sensitivity to partner seed.
Figure 26: Cross-play realism matrices across Pareto-frontier checkpoints (Boston Singapore). Each panel corresponds to one selected checkpoint on the success–realism Pareto frontier. Within each panel, entry reports the ego realism meta score when pairing seed with seed ( total seeds). Diagonal entries are self-play (same seed paired with itself), while off-diagonals are cross-play (different seeds). Off-diagonal performance is comparable to diagonal performance across checkpoints, indicating minimal sensitivity to partner seed.
| Checkpoint | Realism Self-Play | Realism Cross-Play | Success Rate Self-Play | Success Rate Cross-Play |
|---|---|---|---|---|
| 4860 | 0.740 0.003 | 0.738 0.003 | 0.751 0.077 | 0.749 0.046 |
| 4910 | 0.740 0.003 | 0.739 0.003 | 0.724 0.036 | 0.721 0.023 |
| 4920 | 0.738 0.004 | 0.737 0.004 | 0.756 0.042 | 0.755 0.029 |
| 4940 | 0.738 0.004 | 0.737 0.002 | 0.776 0.012 | 0.771 0.012 |
| 6180 | 0.737 0.003 | 0.735 0.002 | 0.783 0.089 | 0.783 0.055 |
| 6240 | 0.736 0.005 | 0.735 0.004 | 0.828 0.056 | 0.825 0.035 |
Table 12: Self-play vs cross-play summary at Pareto-frontier checkpoints (Singapore Boston). We evaluate each selected frontier checkpoint under both self-play (paired with the same-seed partner) and cross-play (paired with different-seed partners) using independently trained seeds. We report mean standard deviation across pairings. Cross-play success and realism closely match self-play at the same checkpoint, indicating that NOMAD does not rely on seed-specific conventions.
Figure 27: Cross-play success matrices across Pareto-frontier checkpoints (Singapore Boston). Each panel corresponds to one selected checkpoint on the success–realism Pareto frontier. Within each panel, entry reports the ego success rate when pairing seed with seed ( total seeds). Diagonal entries are self-play (same seed paired with itself), while off-diagonals are cross-play (different seeds). Off-diagonal performance is comparable to diagonal performance across checkpoints, indicating minimal sensitivity to partner seed.
Figure 28: Cross-play realism matrices across Pareto-frontier checkpoints (Singapore Boston). Each panel corresponds to one selected checkpoint on the success–realism Pareto frontier. Within each panel, entry reports the ego realism meta score when pairing seed with seed ( total seeds). Diagonal entries are self-play (same seed paired with itself), while off-diagonals are cross-play (different seeds). Off-diagonal performance is comparable to diagonal performance across checkpoints, indicating minimal sensitivity to partner seed.
| Checkpoint | Realism Self-Play | Realism Cross-Play | Success Rate Self-Play | Success Rate Cross-Play |
|---|---|---|---|---|
| 6030 | 0.726 0.007 | 0.726 0.005 | 0.809 0.029 | 0.809 0.018 |
| 6040 | 0.730 0.005 | 0.728 0.005 | 0.796 0.024 | 0.793 0.015 |
| 6080 | 0.727 0.006 | 0.727 0.005 | 0.806 0.029 | 0.804 0.016 |
| 6340 | 0.728 0.006 | 0.726 0.004 | 0.798 0.041 | 0.794 0.024 |
| 6550 | 0.727 0.005 | 0.725 0.003 | 0.809 0.025 | 0.812 0.017 |
| 7010 | 0.725 0.004 | 0.724 0.004 | 0.843 0.031 | 0.838 0.018 |
Table 13: Self-play vs cross-play summary at Pareto-frontier checkpoints (Singapore Pittsburgh). We evaluate each selected frontier checkpoint under both self-play (paired with the same-seed partner) and cross-play (paired with different-seed partners) using independently trained seeds. We report mean standard deviation across pairings. Cross-play success and realism closely match self-play at the same checkpoint, indicating that NOMAD does not rely on seed-specific conventions.
Figure 29: Cross-play success matrices across Pareto-frontier checkpoints (Singapore Pittsburgh). Each panel corresponds to one selected checkpoint on the success–realism Pareto frontier. Within each panel, entry reports the ego success rate when pairing seed with seed ( total seeds). Diagonal entries are self-play (same seed paired with itself), while off-diagonals are cross-play (different seeds). Off-diagonal performance is comparable to diagonal performance across checkpoints, indicating minimal sensitivity to partner seed.
Figure 30: Cross-play realism matrices across Pareto-frontier checkpoints (Singapore Pittsburgh). Each panel corresponds to one selected checkpoint on the success–realism Pareto frontier. Within each panel, entry reports the ego realism meta score when pairing seed with seed ( total seeds). Diagonal entries are self-play (same seed paired with itself), while off-diagonals are cross-play (different seeds). Off-diagonal performance is comparable to diagonal performance across checkpoints, indicating minimal sensitivity to partner seed.
Figure 31: Pareto frontier of success rate versus realism meta score across training checkpoints for Boston-to-Pittsburgh transfer. Each point corresponds to a policy checkpoint, colored by the number of interaction steps. The yellow star denotes the zero-shot transfer policy from Boston, while the lime star shows behavior cloning using Pittsburgh trajectories. The red dashed line and points indicate the empirical Pareto frontier on this test set. Frontier checkpoints consistently dominate the baselines and exhibit only a mild trade-off between task success and realism.
| Policy | Scenarios | Realism Meta Score | Success Rate |
|---|---|---|---|
| Original BOS | 0.7022 | 55.77% | |
| Original SIN | 0.6795 | 42.25% | |
| Mirrored SIN | 0.6756 | 43.96% | |
| Original SIN | 0.7011 | 55.49% | |
| Original BOS | 0.6585 | 36.67% | |
| Mirrored BOS | 0.6278 | 38.20% |
Table 14: Effect of map mirroring on zero-shot cross-city transfer. BOS and SIN denote Boston and Singapore scenarios. and denote behavior cloning policies trained in Boston and Singapore. Mirroring the city map yields only modest performance improvements, indicating that cross-city distribution shift extends beyond driving-side conventions and cannot be resolved by geometric transformations alone.
| Dataset | Value |
|---|---|
| Test set@Singapore | 0.8581 |
| New test set@Singapore | 0.8602 |
| Test set@Boston | 0.8311 |
| New test set@Boston | 0.8284 |
| Test set@Pittsburgh | 0.8339 |
| New test set@Pittsburgh | 0.8392 |
Table 15: The WOSAC realism meta scores of human trajectories in different sets of scenarios.
Figure 32: Pareto frontier of success rate versus realism meta score across training checkpoints under Boston-to-Singapore transfer. Each point corresponds to a policy checkpoint, colored by the number of interaction steps. The yellow star denotes the zero-shot transfer policy from Boston, while the lime star shows behavior cloning using Singapore trajectories. The red dashed line and points indicate the empirical Pareto frontier on this test set. Frontier checkpoints consistently dominate the baselines and exhibit only a mild trade-off between task success and realism. Blue triangles report the performance of selected frontier policies on a held-out Singapore test set, confirming that the Pareto dominance generalizes beyond the test data.
Figure 33: Pareto frontier of success rate versus realism meta score across training checkpoints under Singapore-to-Boston transfer. Each point corresponds to a policy checkpoint, colored by the number of interaction steps. The yellow star denotes the zero-shot transfer policy from Singapore, while the lime star shows behavior cloning using Boston trajectories. The red dashed line and points indicate the empirical Pareto frontier on this test set. Frontier checkpoints consistently dominate the baselines and exhibit only a mild trade-off between task success and realism. Blue triangles report the performance of selected frontier policies on a held-out Boston test set, confirming that the Pareto dominance generalizes beyond the test data.
Figure 34: Pareto frontier of success rate versus realism meta score across training checkpoints under Singapore-to-Pittsburgh transfer. Each point corresponds to a policy checkpoint, colored by the number of interaction steps. The yellow star denotes the zero-shot transfer policy from Singapore, while the lime star shows behavior cloning using Pittsburgh trajectories. The red dashed line and points indicate the empirical Pareto frontier on this test set. Frontier checkpoints consistently dominate the baselines and exhibit only a mild trade-off between task success and realism. Blue triangles report the performance of selected frontier policies on a held-out Pittsburgh test set, confirming that the Pareto dominance generalizes beyond the test data.
Figure 35: Overly conservative behavior when facing curbed parking vehicles. When facing many static parking vehicles, the vehicle 33 shows overly conservative behavior and thus fails to reach the goal in time. The gray dashed line shows the path the vehicle should follow and the red one shows the generated trajectory.
Figure 36: Being lost if initialized off-road. The vehicle 9 is initialized outside the road. At the beginning, the vehicle tries to go on road. However, as the goal is outside of the road as well, it finally turn left and drive outside of the road again.
Figure 37: Narrow-lane failure under Boston-to-Singapore transfer. Representative frames from a Boston-to-Singapore transfer scenario illustrating geometric mismatch in narrow roads. (a) Zero-shot transferred policy. (b) NOMAD-adapted policy. The zero-shot policy, trained in Boston, fails to account for the narrower lane geometry commonly found in Singapore: by s vehicles 1 and 2 have both drifted outside the drivable region, and vehicle 1 never reaches its goal. In contrast, NOMAD adapts to the target-city road geometry through map-based self-play and maintains stable lane-centering throughout the maneuver, so both vehicles reach their goals. Numbers identify the controlled vehicles and marks the goal of vehicle ; a controlled vehicle is drawn in orange while it is off-road.
Figure 38: Driving-side mismatch under Boston-to-Singapore transfer. (a) Zero-shot transferred policy. (b) NOMAD-adapted policy. The zero-shot policy exhibits a systematic bias toward right-hand driving learned in the source city: vehicles 1 (moving down) and 2 (moving up) both drift to their right, toward the opposing lane, and collide head-on at s in the left-hand traffic setting of Singapore, and neither vehicle reaches its goal. In contrast, NOMAD adapts its lane-keeping behavior through map-based self-play and consistently drives on the correct side of the road, so both vehicles reach their goals. Numbers identify the controlled vehicles and marks the goal of vehicle ; a controlled vehicle is drawn in red while it is in collision and in orange while it is off-road.
Figure 39: Driving-side mismatch under Singapore-to-Boston transfer (same scenario as Figure 5 ). (a) Zero-shot transferred policy. (b) NOMAD-adapted policy. The zero-shot policy retains a left-hand driving bias learned in the source city: vehicle 2 drifts into the lower lane, which belongs to oncoming traffic, and collides head-on with vehicle 1 at s, and vehicle 4 drifts to its left until it leaves the road, resulting in lane violations, collisions, and off-road behavior in the right-hand traffic setting of Boston. In contrast, NOMAD adapts its lane-keeping behavior via map-based self-play and consistently drives on the correct side of the road, so all four vehicles reach their goals. Numbers identify the controlled vehicles and marks the goal of vehicle ; a controlled vehicle is drawn in red while it is in collision and in orange while it is off-road.
Figure 40: Cascading collision failure from a Boston-to-Singapore transfer scenario involving dense static obstacles. (a) Zero-shot transferred policy. (b) NOMAD-adapted policy. The scene is rotated so that the vehicles of interest travel from left to right. Under the zero-shot policy, vehicle 1 fails to execute an early rightward avoidance maneuver and collides with the first static vehicle at s. Following the collision, it initiates a right-turn corrective action, whereas a leftward maneuver is required to avoid the second obstacle, leading to a second collision at s. In contrast, NOMAD selects appropriate avoidance actions at both decision points, and all three controlled vehicles reach their goals, illustrating improved closed-loop decision making under target-city geometric and obstacle-layout shifts. Numbers identify the controlled vehicles and marks the goal of vehicle ; a controlled vehicle is drawn in red while it is in collision and in orange while it is off-road.
Figure 41: Interaction mismatch frames from a Singapore-to-Boston transfer scenario involving multiple interacting vehicles. (a) Zero-shot transferred policy. (b) NOMAD-adapted policy. The zero-shot policy exhibits miscalibrated interaction behavior: on the upper-left approach, vehicle 2 travels considerably faster than vehicle 3 ahead of it (about 10 vs. 6 m/s) and rear-ends it at s, while at the intersection vehicle 6 enters without yielding to vehicle 4, causing a second collision at s. In contrast, NOMAD learns more balanced speed regulation and yielding behavior through map-based self-play: all six controlled vehicles reach their goals without collision or off-road events. Numbers identify the controlled vehicles and marks the goal of vehicle ; a controlled vehicle is drawn in red while it is in collision and in orange while it is off-road.
More in this topic
Autonomous Driving
Autonomous driving develops vehicles that perceive road scenes, predict interactions, and plan and control safe movement with limited or no human input.
See all 536 papers